Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (V olume 1: Long Papers) , pages 2509–2525
July 27 - August 1, 2025 ©2025 Association for Computational Linguistics
INVESTOR BENCH : A Benchmark for Financial Decision-Making Tasks
with LLM-based Agents
Haohang Li1,⋆, Yupeng Cao1,⋆, Yangyang Yu1,⋆, Shashidhar Reddy Javaji 1,
Zhiyang Deng1, Yueru He2, Yuechen Jiang1, Zining Zhu 1,
Koduvayur Subbalakshmi1, Jimin Huang 4, Lingfei Qian 4, Xueqing Peng 4,
Qianqian Xie4,†, Jordan W. Suchow1
1Stevens Institute of Technology 2Columbia University
3Harvard University 4The Fin AI
⋆Equal Contribution † Corresponding author: qianqian.xie@yale.edu
Abstract
Recent advancements have underscored the po-
tential of large language model (LLM)-based
agents in ﬁnancial decision-making. Despite
this progress, the ﬁeld currently encounters
two main challenges: (1) the lack of a com-
prehensive LLM agent framework adaptable
to a variety of ﬁnancial tasks, and (2) the ab-
sence of standardized benchmarks and consis-
tent datasets for assessing agent performance.
To tackle these issues, we introduce INVESTOR -
BENCH , the ﬁrst benchmark speciﬁcally de-
signed for evaluating LLM-based agents in di-
verse ﬁnancial decision-making contexts. I N-
VESTOR BENCH enhances the versatility of
LLM-enabled agents by providing a compre-
hensive suite of tasks applicable to different ﬁ-
nancial products, including single equities like
stocks, cryptocurrencies and exchange-traded
funds (ETFs). Additionally, we assess the rea-
soning and decision-making capabilities of our
agent framework using thirteen different LLMs
as backbone models, across various market en-
vironments and tasks. Furthermore, we have cu-
rated a diverse collection of open-source, multi-
modal datasets and developed a comprehensive
suite of environments for ﬁnancial decision-
making. This establishes a highly accessible
platform for evaluating ﬁnancial agents’ per-
formance across various scenarios. The code
is available at Github Repo: https://github.
com/felis33/INVESTOR-BENCH
1 Introduction
The recent studies on large language model (LLM)-
based agents have demonstrated impressive perfor-
mance across a range of decision-making tasks in
complex and open-ended environments spanning
various domains ( Zhang et al. , 2024b; Guo et al. ,
2024; Eigner and Händler, 2024; Wang et al., 2024).
However, developing agentic frameworks tailored
speciﬁcally for ﬁnancial decision-making remains a
signiﬁcant challenge. This complexity arises from
Figure 1: General architecture of I NVESTOR BENCH .
the need for agents to acutely discern and priori-
tize decisive signals, and then make sequentially
high-quality decisions within the volatile and multi-
faceted ﬁnancial markets, where information varies
in time sensitivity and modality.
Furthermore, the design of ﬁnancial agents be-
comes increasingly complex when applied across
multiple decision-making tasks, due to the signiﬁ-
cant variation in key factors inﬂuencing ﬁnancial
decisions across different objectives and task types.
For instance, single-equity tasks like stock trading
require analyzing company-speciﬁc and industry-
wide data, including market metrics, sector trends,
performance reports, and relevant news ( Yi et al.,
2022). In contrast, cryptocurrency trading is highly
sensitive to crypto-speciﬁc news and sentiment
due to its dynamic nature ( Bhatnagar et al., 2023).
ETFs, on the other hand, typically follow passive in-
vestment strategies, emphasizing long-term growth
and cost efﬁciency ( Madhavan, 2016).
The recent emergence of ﬁnancial LLM-based
2509

agent frameworks such as F INMEM (Y u et al. ,
2024a), F INAGENT (Zhang et al., 2024a), C RYPTO -
TRADE (Li et al. , 2024), F INROBOT (Y ang et al.,
2024), and F INCON (Y u et al., 2024b) has pre-
sented a variety of architectural approaches tai-
lored to speciﬁc ﬁnancial tasks. This diversiﬁca-
tion has sparked substantial interest across both
academic and industrial landscapes. F INROBOT is
engineered speciﬁcally for market analysis, while
FINMEM and F INAGENT are oriented towards
trading individual equities like stocks and ETFs.
CRYPTO TRADE focuses solely on cryptocurrency
trading. F INCON pioneers in addressing portfolio
management, although it currently handles only
compact portfolios consisting of three stock assets.
While these frameworks are effective within their
respective niches, they generally focus on address-
ing only limited types of ﬁnancial decision-making
tasks. This restricts them from further demonstrat-
ing the broader applicability of these frameworks
and limits the comprehensive, comparative insights
that could be drawn from their overall decision-
making performance. Furthermore, the frequent
reliance on proprietary ﬁnancial data complicates
the evaluation of these tools, obscuring their ef-
fectiveness and adaptability in broader contexts.
Therefore, there is a pressing need to develop inno-
vative benchmarks speciﬁcally designed to evalu-
ate LLM-based agents across a wider spectrum of
ﬁnancial decision-making scenarios. Such bench-
marks would enable a more robust assessment of
these technologies, facilitating advancements that
could cater to various ﬁnancial applications.
We introduce I NVESTOR BENCH , an open-
source, LLM-based agent benchmark that gener-
alizes across a broad range of ﬁnancial decision-
making tasks. Its detailed structure is illustrated
in Figure 1. Further developed upon the founda-
tional framework of F INMEM (Y u et al., 2024a),
which focuses on single-stock investment decisions,
our benchmark extends the scope to encompass
an ensemble of diverse ﬁnancial market envi-
ronments for various ﬁnancial tasks. I NVESTOR -
BENCH ’s cognitive architecture, similar to F IN-
MEM, employs a layered memory processing
mechanism with distinct decay rates, enabling the
agent to store, retrieve, and consolidate insights and
reﬂections more effectively than the pure similarity-
based memory retrieval used in F INAGENT . This
approach ensures that decisions are informed by
timely and impactful data, a capability previously
shown effective for single-asset trading. These fea-
tures reﬂect how human traders draw sequential
decisions upon investment signals from multiple
sources and varying time sensitivities, allowing the
agent to naturally adapt to complex ﬁnancial tasks.
INVESTOR BENCH expands its evaluation beyond
the original stock trading tasks to encompass three
decision tasks signiﬁcant in the realm of ﬁnancial
investment: stock trading, cryptocurrency trad-
ing, and ETF investing.
In summary, we make three key contributions:
1) We establish I NVESTOR BENCH , an innovative
and comprehensive ﬁnancial agentic benchmark
designed to evaluate the reasoning and sequen-
tial decision-making capabilities of LLM-based
agents in complex, open-ended ﬁnancial scenar-
ios. This benchmark provides a realistic perspec-
tive for assessing the design and performance of
such agents. 2) We provide a set of open-source,
multi-source market environments that closely
mirror real-world conditions. Furthermore, these
environments also serve as a standardized platform
for evaluating the decision-making performance of
other LLM-based ﬁnancial agents. 3) We present a
uniﬁed, ﬂexible language-agent framework that
allows ﬁnance professionals to conveniently cus-
tomize assess any LLMs serving as the agents rea-
soning core. In this paper, we conduct a holistic
evaluation of 13 LLMs including recent, compet-
itive, and domain-speciﬁc ﬁne-tuned models (see
Table 1) to provide a broad overview of their rea-
soning capabilities in sequential decision-making
tasks within ﬁnancial contexts.
2 LLM Trading Agents
In this section, we deﬁne a framework of the LLM-
based agents in the I NVESTOR BENCH and formal-
ize the ﬁnancial decision-making tasks within the
context of partially observable Markov decision
process (POMDP) ( Bertsekas and Shreve , 1996;
Liu et al., 2020; Kabbani and Duman, 2022).
2.1 Deﬁnition
The LLM-based agent in I NVESTOR BENCH is
structured as a large language model-modulo frame-
work, designed to match or surpass the capabilities
of professional human investors. This framework
consists of several interconnected modules, each
tailored to handle the distinct challenges presented
by the ﬁnancial markets volatility and complexity:
Brain/Backbone (LLM): This module, which is
the LLM itself, serves as the core of the LLM-
2510

based agent. It enhances the agent’s capabilities
by enabling it to understand, process, and generate
natural language. This module plays a crucial role
in supporting complex decision-making processes,
offering interpretations of market-related informa-
tion, generating predictive analytics, and reﬂecting
on past investment decisions.
Perception: This module serves a critical function
by converting raw market data into a structured
format that is compatible with the LLM, specifying
what the agent perceives and observes, which
includes numerical, textual, and visual information.
Proﬁle: This module serves two functions
articulated in natural language. Firstly, it describes
the agent’s role, highlighting its character as an
experienced investor with expert-level knowledge
and a self-adaptive risk preference. This risk
preference dynamically adjusts based on historical
market momentum, allowing the agent to optimize
its strategies in real time. Secondly, the module
provides a detailed background of the decision-
making task, specifying the key characteristics
and pertinent information about the target assets
involved in the trading decisions, such as equity
historical performance, price ﬂuctuations, and
sector information. This dual-function module
supports the agent’s decisions with both the current
market context and its historical performance.
Memory: This module processes and retains es-
sential market data and historical insights, allowing
the agent to draw on a rich repository of knowledge
for decision-making. Building upon the pioneer-
ing work of Y u et al. (2024a) in F INMEM, the
memory architecture comprises two primary com-
ponents: Working Memory and Layered Long-
Term Memory, as depicted in Figure 2.
Working memory: This component maintains
FINMEM’s original functionalities: observation,
summarization, and reﬂection. It incorporates two
reﬂection mechanisms: immediate and extended.
Immediate reﬂection produces the agent’s reason-
ing outcomes by integrating current market indi-
cators with the top- K ranked events from each
long-term memory layer, which are signiﬁcant dur-
ing both warm-up and evaluation stages. In the
warm-up stage, the emphasis shifts as the trading
direction is predetermined, focusing on understand-
ing market trends and enhancing predictive accu-
racy. In the evaluation stage, it outputs the trading
direction (Buy, Sell, or Hold), the rationale for this
decision, identifying the most inﬂuential memory
events and their respective IDs from each layer.
Layered Long-Term Memory: Inspired by the
human cognitive system’s varying information de-
cay speeds, Layered Long-Term Memory com-
ponent structures ﬁnancial insights across multi-
ple layers. Each layer is represented by a vector
database in the Long-Term Memory data ware-
house, where information is prioritized and purged
based on a speciﬁc decay rate. Deeper layers retain
information longer with smaller decay rates, while
shallower layers, dealing with more transient data,
have larger decay rates. This tiered approach is
critical as it allows the adaptation of the memory
architecture to a broader range of ﬁnancial tasks
beyond single-asset decisions, accommodating an
expanded variety of data sources and increasing
overall system ﬂexibility. Detailed mechanisms for
ranking and decay in each layer are further elabo-
rated in the Appendix A.
Action: This module executes trading and in-
vestment decisions based on the analysis pro-
vided by other modules. It directly outputs
{“Buy", “Sell", “Hold" } for traded asset (stock,
crypto, or ETF), as instructed by the backbone
LLM. Action module synthesizes the operational
outcomes from the Proﬁle and Memory modules
to facilitate precise and well-informed investment
decisions. For its daily trading operations, the
agent can choose from three speciﬁc actions for
the traded asset: “Buy", “Sell", or “Hold". The
functionality and input requirements of this module
differ signiﬁcantly between the warm-up and eval-
uation stages: during the warm-up stage , the agent
observes daily adjusted price differences between
consecutive days, which are critical for identify-
ing potential “Buy" or “Sell" signals. This period
allows the agent to calibrate and adjust its decision-
making strategies based on near-term market move-
ments; during the evaluation stage, access to future
price data is restricted, compelling the agent to rely
solely on available historical data and its cognitive
processing capabilities. In response to trading in-
quiries, the module integrates historical Proﬁt &
Loss (PnL), outcomes from extended reﬂections,
and the top-K retrieved memories. This integration
ensures that each trading decision is grounded in
a comprehensive analysis of past performance and
current market conditions.
2511

Figure 2: (1) The language agent’s memory module is crafted to interact with the market environment to conduct various
ﬁnancial decision-making tasks. It contains two core components – Working Memory and Layered Long-term Memory. (2) The
outline of the agent’s decision-making workﬂow for retrieving critical memory events and market observations to inform speciﬁc
investment decisions.
2.2 Modeling ﬁnancial decision-making
Formally, we model a ﬁnancial decision-making
process as inﬁnite horizon POMDP with time index
T = {0, 1, 2, · · · } and discount factor α ∈ (0, 1].
This POMDP contains: (1) a state space X × Y
where X is the observable component and Y is un-
observable component of the ﬁnancial market; (2)
the action space of the agent is A, which is mod-
eled as {“Buy", “Sell", “Hold" }; (3) the reward
function R(o, b, a) : X × Y × A → R uses daily
proﬁt & loss (PnL) as the output; (4) the observa-
tion process {Ot}t∈T ⊆ X is a multi-dimensional
process (5) the reﬂection process {Bt}t∈T ⊆ Y
represents the agent’s self-reﬂection, which is up-
dated from Bt to Bt+1 on daily basis ( Grifﬁths
et al., 2023); (6) the action At ∼ π(·|prompt) rep-
resents the way to make investment decision driven
by the language conditioned policy π. By denoting
daily proﬁt & loss (PnLs) by Rπ
t = R(Ot, Bt, At)
and the set of all admissible language conditioned
policies as Π = {π(·|prompt)}, the optimization
objective of ﬁnancial decision-making task is then:
max
π∈Π
E
[ ∑
t∈T
αtRπ
t
]
(1)
3 InvestorBench
He we introduce the detailed architecture of In-
vestorBench, as illustrated in Figure 1.
3.1 Benchmark Composition
INVESTOR BENCH is organized into four main com-
ponents: (1) Data Sources and Market Environ-
ments: I NVESTOR BENCH utilizes a wide range of
open-source data and incorporates third-party APIs,
such as Y ahoo Finance and SEC EDGAR, to create
a comprehensive, multi-modal market environment
data warehouse. (2) LLM Agent : I NVESTOR -
BENCH includes an advanced LLM-based agent
equipped with modules for Brain, Perception, Pro-
ﬁle, Memory, and Action. This agent is enhanced
with external tools (such as tabular data readers and
API callers) and data operations (including vector
database management, information reinforcement,
and retrieval). (3) Financial Decision-Making
Tasks: I NVESTOR BENCH offers three distinct ﬁ-
nancial decision-making tasks, differentiated by
their asset types. (4) Evaluation Metrics: The efﬁ-
cacy of all tasks within I NVESTOR BENCH is evalu-
ated using a set of standard metrics in the quantita-
tive ﬁnance ﬁeld, providing a thorough evaluation
of the decision-making capabilities of the LLM-
based agent.
3.2 Trading Environments
We release three datasets, each curated from
diverse sources, to construct tailored ﬁnancial
market environments for speciﬁc tasks. Our
objective is to address the current gap in evaluation
environments for ﬁnancial decision-making agent
frameworks and to offer a fully open platform
2512

for the comprehensive assessment of agents
across various tasks. Below, we introduce each
environment, categorized by task type, detailing its
scope and the data sources it incorporates.
Stock market environment integrates information
from multiple sources, including: 1) Daily stock
open, high, low, close, and volume (OHLCV) data
acquired from Y ahoo Finance. 2) Summarized
insights from company quarterly and annual
reports (Form 10-Q and 10-K) downloaded from
the SEC EDGAR database. 3) News articles for
seven stocks collected daily between 2020-07-01,
and 2021-05-06. The news data for four of these
companiesMicrosoft Corporation (MSFT), John-
son & Johnson (JNJ), UVV Corporation (UVV),
and Honeywell International Inc. (HON)-are
randomly selected from the pool with the most new
records (over ﬁve hundred) from the open-access
dataset provided by Zhou et al. ( Zhou et al. ,
2021), while the news data for the remaining
three companiesTesla, Inc. (TSLA), Apple Inc.
(AAPL), and NIO Inc. (NIO)-are obtained from
Reﬁnitiv Real-Time News, which primarily
contains high-quality news information from
Reuters. 4) The sentiment categories ( ’positive’,
’negative’, ’neutral’) assigned to each news record
are generated by gpt-3.5-turbo-0125.
Cryptocurrency market environment encom-
passes 1) the daily stock open-high-low-close-
volume (OHLCV) acquired from CoinMarketCap;
2) the multisource cryptocurrency news data
collected from cryptonews, cryptopotato, and
cointelegraph(V anhoucke, 2023); 3) news spanning
from 2023-02-13 to 2023-11-05 collected by
(Zhou et al. , 2021) in daily frequency. 4) The
sentiment categories generated by the same means.
ETF market environment is constructed using
News-Informed Financial Trend Yield (NIFTY)
dataset ( Saqur et al. , 2024). It contains the pro-
cessed and curated daily news headlines from 2019-
07-29 to 2020-09-21 and generated sentiment cate-
gories for each news headline.
In experimental use, we divide the dataset ac-
cording to the date, with the train set used for the
warmup phase to establish the memory database,
and the test set used for the test phase to evaluate
the model performance.
Table 1: I NVESTOR BENCH evaluates 13 proprietary or
open-source LLMs on ﬁnancial decision-making tasks.
Model #Size Form V er.Model #Size Form V er.
gpt-4(Achiam et al., 2023) N/A api 0613 Qwen2.5-7b(Qwen team, 2024) 7B open Instruct
gpt-4o(OpenAI, 2022) N/A api 0806 Qwen2.5-32b(Qwen team, 2024) 32B open Instruct
gpt-o1-preview N/A api 0912 Qwen2.5-72b(Qwen team, 2024) 72B open Instruct
DeepSeek-v2(Xin et al. , 2024) 15B open Lite llama3.1-8b(Llama team, 2024) 8B open Instruct
DeepSeek-llm(Xin et al. , 2024) 67B open Chat llama3.1-70b(Llama team, 2024) 70B open Instruct
Yi-1.5-9b(Y oung et al., 2024) 9B open Chat Palmyra-Fin(team, 2024) 70B open 32K
Yi-1.5-34b(Y oung et al., 2024) 34B open Chat
3.3 Evaluation metrics
We employ four widely recognized ﬁnancial met-
rics to evaluate and compare the investment per-
formance of various LLMs serving as backbones
across different tasks: : Cumulative Return (CR)
(Hull, 2007), Sharpe Ratio (SR) ( Sharpe, 1994),
Annualized V olatility (A V) (Cochrane, 1988), and
Maximum Drawdown(MDD) ( Ang and Chen ,
2003). Note that CR and the SR are often con-
sidered more essential than A V and MDD in evalu-
ating asset trading performance due to their focus
on long-term gains and risk-adjusted returns by
their deﬁnition. Here, we regard these two met-
rics as primary metrics when evaluating the exper-
iment outcomes. The detailed explanation is in
Appendix B.
4 Experiment and Discussion
To establish a baseline and assess the performance
of LLM agents, we standardize experimental set-
tings and evaluation metrics across various ﬁnan-
cial decision-making tasks. Results are presented
on a task-by-task basis. We report the performance
of I NVESTOR BENCH on three single-asset trading
tasks: stocks, cryptocurrencies, and ETFs trad-
ing, using closed-source, open-source, and domain-
speciﬁc LLMs.
4.1 Experiment Setup
Table 1 summarizes the performance of a compre-
hensive list of trading agents. For single equity
tasks, the baseline is set up by Buy and Hold strat-
egy, while for portfolio management task, it is set
up by an equal-weight portfolio with the detailed
rational explained in Appendix. In our experiments,
the temperature parameter of all LLM-based agent
systems is set at 0.6 to balance response consis-
tency and reasoning creativity. The performance
metrics are reported for the test trajectory with the
median CR, SR, A V , and MDD from ﬁve repeated
epochs. (If the median of these metrics does not
belong to the same epoch, the performance is based
on the trajectory with the median SR.)
Furthermore, the selection of warm-up and test
periods differs across various tasks due to the vary-
2513

Table 2: Performance of stock trading with different LLMs as backbone model across seven stocks.
Model MSFT JNJ UVV HON
CR↑ SR↑ A V↓ MDD↓ CR↑ SR↑ A V↓ MDD↓ CR↑ SR↑ A V↓ MDD↓ CR↑ SR↑ A V↓ MDD↓
Buy & Hold 15.340 0.717 17.236 9.428 13.895 0.927 12.075 9.847 36.583 1.457 20.216 15.406 33.256 1.619 16.537 9.195
Financial Domain Models
Palmyra-Fin-70B 14.697 0.618 18.987 9.428 5.748 0.311 13.329 9.367 37.875 1.407 21.528 15.967 20.016 1.010 15.852 6.824
Proprietary Models
GPT-o1-preview 17.184 0.664 20.700 9.428 13.561 0.749 14.396 9.847 41.508 1.481 22.411 9.633 13.162 0.535 19.673 11.558
GPT-4 16.654 0.643 20.715 9.428 13.712 0.761 14.417 9.860 31.791 1.132 22.471 10.434 34.342 1.383 19.858 9.195
GPT-4o 12.461 0.638 15.631 6.647 9.099 0.604 12.055 7.169 8.043 0.342 18.796 14.889 38.540 1.668 18.480 8.979
Open-Source Models
Qwen2.5-72B-Instruct 7.421 0.406 14.654 6.973 14.353 0.787 14.487 9.812 37.178 1.257 23.614 13.365 34.309 1.380 19.858 9.292
Llama-3.1-70B-Instruct 17.396 0.921 15.105 7.045 13.868 0.773 14.338 9.825 35.981 1.192 24.140 15.406 43.944 1.826 19.253 8.993
DeepSeek-67B-Chat 13.941 0.575 19.376 7.850 14.426 0.818 14.111 9.825 29.940 1.022 23.435 15.407 32.536 1.317 19.753 10.782
Yi-1.5-34B-Chat 22.093 0.865 20.433 9.428 14.004 0.814 13.757 9.847 20.889 0.704 23.748 14.936 30.743 1.258 19.551 9.195
Qwen2.5-32B-Instruct -0.557 -0.028 15.796 8.946 2.905 0.201 11.540 7.169 -1.623 -0.067 19.301 17.986 26.332 1.366 15.420 5.261
DeepSeek-V2-Lite (15.7B) 11.904 0.479 19.869 16.094 -7.482 -0.462 12.953 17.806 33.560 1.175 22.838 12.984 16.686 0.672 19.852 16.806
Yi-1.5-9B-Chat 19.333 0.755 20.486 9.428 18.606 1.112 13.392 10.986 49.415 1.663 23.768 11.430 29.028 1.173 19.791 12.588
Llama-3.1-8B-Instruct 22.703 0.912 19.910 7.385 13.988 1.025 14.117 9.969 41.108 1.367 24.057 16.429 39.079 1.601 19.526 10.341
Qwen-2.5-Instruct-7B -10.305 -0.500 16.517 23.371 21.852 0.676 25.823 9.573 11.752 0.588 15.862 15.451 4.291 0.197 17.204 14.156
Model TSLA AAPL NIO A verage
CR↑ SR↑ A V↓ MDD↓ CR↑ SR↑ A V↓ MDD↓ CR↑ SR↑ A V↓ MDD↓ CR↑ SR↑ A V↓ MDD↓
Buy & Hold 39.244 0.600 52.339 37.975 10.837 0.324 26.899 19.119 52.216 0.592 73.966 47.766 34.099 0.505 51.068 34.953
Financial Domain Models
Palmyra-Fin-70B -6.661 -0.153 34.761 25.820 8.562 0.256 26.835 25.466 -3.261 -0.039 70.181 58.406 -0.453 0.021 43.925 36.564
Proprietary Models
GPT-o1-preview 34.499 0.549 50.247 35.490 8.238 0.291 22.801 14.412 32.433 0.385 70.704 54.016 25.057 0.408 47.918 34.639
GPT-4 45.246 0.821 44.088 25.031 9.889 0.304 26.226 19.119 75.952 0.887 71.801 37.867 43.696 0.671 47.371 27.339
GPT-4o 45.946 0.930 39.524 21.631 7.405 0.315 18.929 12.824 63.743 0.909 58.795 29.220 39.031 0.718 39.083 21.225
Open-Source Models
Qwen2.5-72B-Instruct 39.112 0.742 42.184 26.985 11.935 0.395 24.352 19.119 87.412 1.505 48.733 12.464 46.153 0.880 38.423 19.523
Llama-3.1-70B-Instruct 37.545 0.615 48.862 29.813 12.772 0.402 25.569 16.021 66.522 0.771 72.345 46.379 38.946 0.596 48.926 30.738
DeepSeek-67B-Chat 35.647 0.611 46.685 33.359 14.213 0.460 24.921 10.876 30.963 0.413 62.891 45.855 26.941 0.495 44.833 30.030
Yi-1.5-34B-Chat 35.364 0.558 50.757 35.490 14.227 0.430 26.631 19.432 64.307 0.733 73.552 48.042 37.966 0.573 50.313 34.321
Qwen2.5-32B-Instruct 21.336 0.503 33.918 20.704 13.220 0.751 14.160 8.943 28.096 0.450 49.917 37.975 20.884 0.568 32.665 22.541
DeepSeek-V2-Lite (15.7B) 31.458 0.513 47.282 35.404 27.016 0.842 8.183 37.435 27.762 0.327 71.203 48.478 28.745 0.561 42.222 40.439
Yi-1.5-9B-Chat 31.350 0.485 51.677 37.975 3.640 0.112 26.183 17.578 33.748 0.393 72.106 55.284 22.913 0.330 49.989 36.946
Llama-3.1-8B-Instruct 35.622 0.574 49.636 36.383 7.079 0.213 26.742 18.747 33.689 0.386 73.205 56.527 25.463 0.391 49.861 37.219
Qwen-2.5-Instruct-7B 41.203 0.638 51.655 37.975 14.336 0.479 24.098 12.029 33.007 0.377 73.442 53.054 29.515 0.498 49.731 34.353
1 The Buy & Hold strategy is a passive investment approach commonly used as a baseline strategy, where an investor
purchases stocks and holds onto them for an extended period regardless of market ﬂuctuations.
2 An upward arrow ( ↑) next to a metric indicates that higher values signify better performance, while a downward arrow ( ↓)
indicates that lower values are preferable.
3 The numbers highlighted in red indicate the best-performing outcomes for the corresponding metrics.
ing time spans of data collected to construct the
agent environment. For the single-asset trading
tasks, the warm-up period of stock trading is from
2020-07-01 to 2020-09-30 and the test period is
from 2020-10-01 to 2021-05-06. The warm-up pe-
riod of cryptocurrency trading is from 2023-02-11
to 2023-04-04 and the test period is from 2023-04-
05 to 2023-11-05. The warm-up period of ETF
trading is from 2019-07-29 to 2019-12-30 and the
test period is from 2020-01-02 to 2020-09-21.
For LLM deployment, we utilize vllm to deploy
LLMs. For small-scale LLMs (under 10B parame-
ters), we deploy models on two RTX A6000 GPUs,
each with 48GB DRAM. For mid-scale LLMs
(10B to 65B parameters), we use four RTX A6000
GPUs. For large-scale LLMs (over 65B parame-
ters), models are deployed on eight A100 GPUs,
each equipped with 80GB DRAM.
4.2 Result 1: Stock Trading
Table 2 presents the performance of thirteen back-
bone models across seven stocks, accompanied by
the average of each metric for all stocks to offer a
more comprehensive view of their overall perfor-
mance. We outline three key insights as follows:
Superior stock trading performance is achieved
with proprietary LLMs as agent backbones
Compared to agents employing open-source or
ﬁnancial-domain-speciﬁc ﬁne-tuned LLMs, those
using the three proprietary LLMs demonstrated sig-
niﬁcantly higher and more consistent average CR
and SR, as shown in Figure 3a. Despite being ﬁne-
tuned with extensive ﬁnancial contexts, domain-
speciﬁc LLMs did not provide a decisive advantage
in sequential stock trading decision-making tasks.
This may be attributed to their primary training for
other functions, such as long ﬁnancial report anal-
ysis exempliﬁed by Palmyra-Fin-70B, rather than
decision-making.
Model parameter size increment enhances agent
ﬁnancial decision-making quality and robust-
ness. In the category of open-source LLMs, those
exceeding 67B parameters displayed superior CRs
and SRs, along with markedly less variance within
their category, as illustrated in Figure 3b and Ta-
ble 2. This underscores the prevailing belief that the
reasoning capabilities of LLMs are proportionate to
their parameter size, which holds also true in stock
2514

(a) The performance comparison by different model types.
(b) By model parameter sizes across open-source LLMs.
Note: Small-size models refer to models with no more than
10B parameters. Medium-size models refer to models with
parameter numbers in the range of (10B, 65B]. Large-size
models are those with more than 65B parameters.
Figure 3: Agent Performance Comparisons from two
key perspectives. The CR, SR, A V , and MDD represent the
average values for each model type, expressed as a percentage
relative to the Buy & Hold strategy.
trading, which is a sequential decision-making task
in an open-ended, volatile environment by nature.
Proprietary models exhibit signiﬁcantly
stronger decision-making capabilities compared
to even the largest open-source LLMs under
complex, mixed market conditions, though this
advantage is less evident in relatively monotone
market environments. During the test phase,
primarily inﬂuenced by the range of open-source
data collected, TSLA and NIO exhibited volatility
with mixed upward and downward stock price
trends, whereas the other ﬁve stocks generally
showed bullish trends. The investment signals
derived from such complex markets tend to be
noisy or delayed, as illustrated in Appendix C. We
observed that proprietary models possess a superior
ability to manage these challenging conditions and
consistently deliver better performance outcomes
than large-sized open-source LLMs. Their
reasoning capability enables them to effectively
utilize other decision-relevant information, such as
historical momentum, current holdings, and, most
critically, self-reﬂection outcomes from the agents,
Table 3: Performance of cryptocurrency trading with dif-
ferent LLMs as backbone models across Bitcoin (BTC)
and Ethereum (ETH).
Model BTC ETH
CR↑ SR↑ A V↓ MDD↓ CR↑ SR↑ A V↓ MDD↓
Buy & Hold 21.821 0.989 54.193 20.796 4.528 0.211 60.551 29.889
Financial Domain Models
Palmyra-Fin-70B -20.812 -1.755 29.012 27.782 4.795 0.348 38.986 16.405
Proprietary Models
GPT-o1-preview 34.060 1.613 51.905 17.075 2.496 0.123 57.400 27.692
GPT-4 22.396 1.199 45.900 17.206 1.516 0.074 57.648 32.541
GPT-4o 14.330 0.770 45.328 17.278 4.666 0.275 47.858 22.539
A verage 23.595 1.195 47.712 17.186 2.893 0.158 54.301 27.591
Open-Source Models
Qwen2.5-72B-Instruct 0.549 0.471 2.866 0.897 11.984 0.846 26.866 27.642
Llama-3.1-70B-Instruct 20.440 1.098 45.763 17.813 -11.888 -0.594 56.540 36.416
DeepSeek-67B-Chat 28.307 1.290 53.893 17.944 9.480 0.447 59.902 26.261
Yi-1.5-34B-Chat 13.620 0.628 53.255 22.790 6.325 0.329 54.304 25.707
Qwen2.5-32B-Instruc 11.566 1.258 22.600 7.984 2.823 0.281 28.339 7.883
DeepSeek-V2-Lite (15.7B) 4.804 0.222 53.353 20.562 -9.504 -0.450 59.656 21.270
Yi-1.5-9B-Chat 7.953 0.366 53.285 26.545 -3.684 -0.172 60.552 35.417
Llama-3.1-8B-Instruct 20.521 0.935 53.924 21.104 4.939 0.236 59.264 29.466
Qwen-2.5-Instruct-7B 19.477 0.886 53.994 20.796 -1.339 -0.109 34.932 -16.053
A verage 14.137 0.795 43.659 17.382 1.015 0.090 48.928 21.557
thereby facilitating more accurate decisions.
4.3 Result 2 & 3: Cryptocurrency Trading
and ETF Trading
In the test phases of both cryptocurrency and ETF
trading tasks, market trends are mixed. Notably,
the cryptocurrency task shows signiﬁcantly smaller
price ﬂuctuations compared to the ETF task. We
outline the key features of using an LLM-agent to
make ﬁnancial decisions across these two distinct
markets as follows:
Large-sized open-source models and propri-
etary models are needed to effectively cap-
ture trading signals of cryptocurrency markets,
which are highly sensitive to news and ﬁnan-
cial sentiment. As shown in Table 3, using mid-
sized and small-sized open-source models as the
decision-making agent backbone generally results
in weaker performance than the market baseline
with respect to CR and SR.
ETF investment requires proprietary models en-
riched with extensive pre-trained knowledge to
serve as the agents brain and provide robust
reasoning support. As shown in Table 4, propri-
etary models signiﬁcantly outperform open-source
and ﬁnancial domain-speciﬁc models in this task.
This advantage arises from the complexity of ETF
trading, which necessitates interpreting actionable
signals across diverse sectors, demanding more
strategic, long-term decisions grounded in deep
comprehension and reﬂection anchored by rich pre-
contexts.
4.4 Discussion
Combining all the experimental results, we ﬁnd
that the performance of different LLM varies sig-
2515

Table 4: Performance of ETF trading with different
LLMs as backbone models.
ETF CR ↑ SR↑ A V↓ MDD↓
Buy & Hold 2.069 0.06 46.645 35.746
Financial Domain Models
Palmyra-Fin-70B 24.759 1.152 30.419 8.203
Proprietary Models
GPT-o1-preview 21.224 0.849 43.766 20.054
GPT-4 2.807 0.110 44.679 37.785
GPT-4o 12.292 0.377 46.150 32.678
A verage 12.108 0.445 44.865 30.172
Open-Source Models
Qwen2.5-72B-Instruct 4.507 0.227 28.090 8.580
Llama-3.1-70B-Instruct 9.895 0.464 30.184 12.759
Yi-1.5-34B-Chat 4.996 0.322 21.986 12.858
Qwen2.5-32B-Instruct 19.617 0.955 29.070 7.496
DeepSeek-V2-Lite (15.7B) 1.389 0.063 31.371 31.831
Yi-1.5-9B-Chat -4.657 -0.228 28.907 15.545
Llama-3.1-8B-Instruct 11.239 0.475 33.480 15.587
Qwen-2.5-Instruct-7B -0.384 -0.020 27.596 14.059
A verage 5.825 0.282 28.835 14.839
niﬁcantly in stock, cryptocurrency, and ETF trad-
ing. This variation not only reﬂects the inherent
complexity of ﬁnancial markets, but also highlights
the importance of model selection or ﬁne-tuning.
For instance, proprietary LLM generally exhibit be
performance in stock trading due to their strong
training on various ﬁnancial datasets, while open-
source models struggle to achieve these results,
especially in more volatile environments such as
cryptocurrency trading. In addition, the effective-
ness of LLM-based agents depends heavily on their
ability to adapt to market ﬂuctuations. Agents
that incorporate advanced memory systems and dy-
namic risk assessment capabilities are better able to
cope with complex market situations, highlighting
the value of the complex architectural features of
LLM-based agent framework in ﬁnancial decision-
making tasks.
5 Related Work
5.1 LLM for Financial Domain
The rapid developement of general-domain lan-
guage models (LMs) has stimulated the explo-
ration of ﬁnancial LMs, such as pre-trained LMs:
FINBERT (Liu et al. , 2021; Y ang et al. , 2020;
Araci, 2019; Huang et al. , 2023), F INBERT-MRC
(Zhang and Zhang , 2023), FLANG ( Shah et al. ,
2022), and several ﬁnancial LLMs: F INGPT( Liu
et al. , 2023), F INMA ( Xie et al. , 2023), I N-
VEST LM ( Y ang et al., 2023), BloombergGPT (Wu
et al., 2023), which leverage extensive training on
diverse ﬁnancial datasets (e.g. stock price data,
ﬁnancial news and analyst reports) and adapt the
capabilities of LMs to the unique needs of ﬁnancial
applications. Concurrently, the advancement of
LLMs has signiﬁcantly enhanced the development
of language-based agent frameworks in the ﬁnan-
cial sector, such as F INMEM (Y u et al., 2024a),
FINAGENT (Zhang et al., 2024a) and F INROBOT
(Y ang et al., 2024), characterized by their adapt-
ability and openness. However, variations in frame-
work design, task scope, and data types present
challenges in uniformly evaluating the efﬁcacy of
LLM agents in ﬁnancial scenarios.
5.2 Financial LLM Benchmarks
In the realm of ﬁnancial LLMs, several benchmarks
have been developed: FLUE ( Shah et al. , 2022) in-
troduces the ﬁrst comprehensive benchmark with
ﬁve ﬁnancial NLP tasks, including sentiment anal-
ysis, headline classiﬁcation, named entity recog-
nition, structure boundary detection, and question
answering. Pixiu ( Xie et al. , 2023) expands this
benchmark to include ﬁnancial document under-
standing and classiﬁcation tasks, incorporating mul-
timodal datasets. FinBen ( Xie et al. , 2024) encom-
passes 36 datasets covering 24 ﬁnancial tasks. De-
spite these advancements, there remains a notable
gap in benchmarks speciﬁcally designed for LLM-
based agent applications within the ﬁnancial sector.
6 Conclusion
INVESTOR BENCH offers the community two dis-
tinct modes of engagement. The ﬁrst mode allows
participants to integrate their ﬁne-tuned LLMs into
the I NVESTOR BENCH ’s agent framework to under-
take ﬁnancial decision-making tasks. This setup
enables them to benchmark the performance of
their models against those previously experimented
with by our work. The second mode permits users
to directly incorporate the environment and evalu-
ation metrics of I NVESTOR BENCH into their own
designed agents, facilitating a comparative analy-
sis of their agent design’s effectiveness. This dual
approach provides a ﬂexible framework for testing
and enhancing ﬁnancial decision-making strategies
within the I NVESTOR BENCH ecosystem.
Future research efforts will expand the bench-
mark by incorporating additional information
modalities, such as audio (e.g., earnings call record-
ings) and graphs (e.g., K-lines, trade charts), to
explore whether these data types can enhance
decision-making quality. The foundational agent
2516

framework of I NVESTOR BENCH is designed to
seamlessly accommodate these modalities, ensur-
ing that the extended benchmark remains easy to
use and scalable.
Limitation
First, I NVESTOR BENCH is currently focusing on
single-asset ﬁnancial decision-making task, with-
out addressing multi-asset tasks such as portfolio
management. Second, copyright restrictions on ﬁ-
nancial domain data may compromise the quality
of the datasets we create, potentially limiting the
assessment of model performance.
Ethical Statement
The authors take full responsibility for the devel-
opment of I NVESTOR BENCH , ensuring that the
publicly available part in dataset does not contain
personal information, and conform to established
ethical guidelines. The data are shared under the
MIT license, requiring users to adhere to its terms.
INVESTOR BENCH is intended for academic and ed-
ucational purposes only and is not a substitute for
professional advice. While efforts have been made
to ensure its accuracy, the authors and their insti-
tutions disclaim liability for any outcomes arising
from its use. Users agree to take responsibility for
ethical and lawful use and to indemnify the authors
and their afﬁliates against any claims or damages
resulting from reliance on this Material.
References
Josh Achiam, Steven Adler, Sandhini Agarwal, Lama
Ahmad, Ilge Akkaya, Florencia Leoni Aleman,
Diogo Almeida, Janko Altenschmidt, Sam Altman,
Shyamal Anadkat, et al. 2023. Gpt-4 technical report.
arXiv preprint arXiv:2303.08774.
Andrew Ang and Joseph Chen. 2003. Downside risk.
Journal of Portfolio Management, 29(4):103–112.
Dogu Araci. 2019. Finbert: Financial sentiment analy-
sis with pre-trained language models. arXiv preprint
arXiv:1908.10063.
Dimitri Bertsekas and Steven E Shreve. 1996. Stochas-
tic optimal control: the discrete-time case , volume 5.
Athena Scientiﬁc.
Mukul Bhatnagar, Sanjay Taneja, and Ramona Rupeika-
Apoga. 2023. Demystifying the effect of the news
(shocks) on crypto market volatility. Journal of Risk
and Financial Management, 16(2):136.
John H. Cochrane. 1988. V olatility tests and efﬁcient
markets: A review essay. Journal of Monetary Eco-
nomics, 22(3):463–485.
Eva Eigner and Thorsten Händler. 2024. Determinants
of llm-assisted decision-making. arXiv preprint
arXiv:2402.17385.
Thomas L Grifﬁths, Jian-Qiao Zhu, Erin Grant, and
R Thomas McCoy. 2023. Bayes in the age of intelli-
gent machines. arXiv preprint arXiv:2311.10206.
Taicheng Guo, Xiuying Chen, Y aqi Wang, Ruidi Chang,
Shichao Pei, Nitesh V Chawla, Olaf Wiest, and Xi-
angliang Zhang. 2024. Large language model based
multi-agents: A survey of progress and challenges.
arXiv preprint arXiv:2402.01680.
Allen H Huang, Hui Wang, and Yi Y ang. 2023. Finbert:
A large language model for extracting information
from ﬁnancial text. Contemporary Accounting Re-
search, 40(2):806–841.
John Hull. 2007. Risk Management and Financial Insti-
tutions. John Wiley & Sons.
Taylan Kabbani and Ekrem Duman. 2022. Deep rein-
forcement learning approach for trading automation
in the stock market. IEEE Access, 10:93564–93574.
Y uan Li, Bingqiao Luo, Qian Wang, Nuo Chen, Xu Liu,
and Bingsheng He. 2024. Cryptotrade: A reﬂective
llm-based agent to guide zero-shot cryptocurrency
trading. In Proceedings of the 2024 Conference on
Empirical Methods in Natural Language Processing ,
pages 1094–1106.
Xiao-Y ang Liu, Guoxuan Wang, and Daochen Zha.
2023. Fingpt: Democratizing internet-scale data
for ﬁnancial large language models. arXiv preprint
arXiv:2307.10485.
Y ang Liu, Qi Liu, Hongke Zhao, Zhen Pan, and Chuan-
ren Liu. 2020. Adaptive quantitative trading: An
imitative deep reinforcement learning approach. In
Proceedings of the AAAI conference on artiﬁcial in-
telligence, volume 34, pages 2128–2135.
Zhuang Liu, Degen Huang, Kaiyu Huang, Zhuang Li,
and Jun Zhao. 2021. Finbert: A pre-trained ﬁnan-
cial language representation model for ﬁnancial text
mining. In Proceedings of the twenty-ninth interna-
tional conference on international joint conferences
on artiﬁcial intelligence, pages 4513–4519.
Meta Llama team. 2024. The llama 3 herd of models .
Ananth N Madhavan. 2016. Exchange-traded funds and
the new dynamics of investing . Oxford University
Press.
Jaap MJ Murre and Joeri Dros. 2015. Replication and
analysis of ebbinghaus forgetting curve. PloS one,
10(7):e0120644.
OpenAI. 2022. Introducing chatgpt.
2517

Joon Sung Park, Joseph O’Brien, Carrie Jun Cai, Mered-
ith Ringel Morris, Percy Liang, and Michael S. Bern-
stein. 2023. Generative agents: Interactive simulacra
of human behavior . In Proceedings of the 36th An-
nual ACM Symposium on User Interface Software
and Technology, UIST ’23, New Y ork, NY , USA.
Association for Computing Machinery.
Alibaba Group Qwen team. 2024. Qwen2.5.
Raeid Saqur, Ken Kato, Nicholas Vinden, and Frank
Rudzicz. 2024. Nifty ﬁnancial news headlines
dataset. arXiv preprint arXiv:2405.09747.
Raj Sanjay Shah, Kunal Chawla, Dheeraj Eidnani,
Agam Shah, Wendi Du, Sudheer Chava, Natraj Ra-
man, Charese Smiley, Jiaao Chen, and Diyi Y ang.
2022. When ﬂue meets ﬂang: Benchmarks and
large pre-trained language model for ﬁnancial do-
main. arXiv preprint arXiv:2211.00083.
William F. Sharpe. 1994. The sharpe ratio. The Journal
of Portfolio Management, 21(1):49–58.
Writer Engineering team. 2024. Palmyra-Fin-70B-32k:
a powerful LLM designed for Finance. https://
dev.writer.com.
Olivier V anhoucke. 2023. Crypto news dataset . Ac-
cessed: 2024-08-20.
Lei Wang, Chen Ma, Xueyang Feng, Zeyu Zhang, Hao
Y ang, Jingsen Zhang, Zhiyuan Chen, Jiakai Tang,
Xu Chen, Y ankai Lin, et al. 2024. A survey on large
language model based autonomous agents. Frontiers
of Computer Science, 18(6):186345.
Shijie Wu, Ozan Irsoy, Steven Lu, V adim Dabravolski,
Mark Dredze, Sebastian Gehrmann, Prabhanjan Kam-
badur, David Rosenberg, and Gideon Mann. 2023.
Bloomberggpt: A large language model for ﬁnance.
arXiv preprint arXiv:2303.17564.
Qianqian Xie, Weiguang Han, Zhengyu Chen, Ruoyu
Xiang, Xiao Zhang, Y ueru He, Mengxi Xiao, Dong
Li, Y ongfu Dai, Duanyu Feng, et al. 2024. The ﬁn-
ben: An holistic ﬁnancial benchmark for large lan-
guage models. arXiv preprint arXiv:2402.12659.
Qianqian Xie, Weiguang Han, Xiao Zhang, Y anzhao
Lai, Min Peng, Alejandro Lopez-Lira, and Jimin
Huang. 2023. Pixiu: A large language model, in-
struction data and evaluation benchmark for ﬁnance.
arXiv preprint arXiv:2306.05443.
Huajian Xin, Z. Z. Ren, Junxiao Song, Zhihong Shao,
Wanjia Zhao, Haocheng Wang, Bo Liu, Liyue Zhang,
Xuan Lu, Qiushi Du, Wenjun Gao, Qihao Zhu, De-
jian Y ang, Zhibin Gou, Z. F. Wu, Fuli Luo, and
Chong Ruan. 2024. Deepseek-prover-v1.5: Harness-
ing proof assistant feedback for reinforcement learn-
ing and monte-carlo tree search .
Hongyang Y ang, Boyu Zhang, Neng Wang, Cheng Guo,
Xiaoli Zhang, Likun Lin, Junlin Wang, Tianyu Zhou,
Mao Guan, Runjia Zhang, et al. 2024. Finrobot: An
open-source ai agent platform for ﬁnancial applica-
tions using large language models. arXiv preprint
arXiv:2405.14767.
Yi Y ang, Yixuan Tang, and Kar Y an Tam. 2023. In-
vestlm: A large language model for investment using
ﬁnancial domain instruction tuning. arXiv preprint
arXiv:2309.13064.
Yi Y ang, Mark Christopher Siy Uy, and Allen Huang.
2020. Finbert: A pretrained language model
for ﬁnancial communications. arXiv preprint
arXiv:2006.08097.
Ziruo Yi, Ting Xiao, Kaz-Onyeakazi Ijeoma, Rat-
nam Cheran, Y uvraj Baweja, and Phillip Nelson.
2022. Stock2vec: An embedding to improve
predictive models for companies. arXiv preprint
arXiv:2201.11290.
Alex Y oung, Bei Chen, Chao Li, Chengen Huang,
Ge Zhang, Guanwei Zhang, Heng Li, Jiangcheng
Zhu, Jianqun Chen, Jing Chang, et al. 2024. Yi:
Open foundation models by 01. ai. arXiv preprint
arXiv:2403.04652.
Y angyang Y u, Haohang Li, Zhi Chen, Y uechen Jiang,
Y ang Li, Denghui Zhang, Rong Liu, Jordan W Su-
chow, and Khaldoun Khashanah. 2024a. Finmem: A
performance-enhanced llm trading agent with layered
memory and character design. In Proceedings of the
AAAI Symposium Series, volume 3, pages 595–597.
Y angyang Y u, Zhiyuan Y ao, Haohang Li, Zhiyang Deng,
Y upeng Cao, Zhi Chen, Jordan W Suchow, Rong Liu,
Zhenyu Cui, Zhaozhuo Xu, et al. 2024b. Fincon: A
synthesized llm multi-agent system with conceptual
verbal reinforcement for enhanced ﬁnancial decision
making. arXiv preprint arXiv:2407.06567.
Wentao Zhang, Lingxuan Zhao, Haochong Xia, Shuo
Sun, Jiaze Sun, Molei Qin, Xinyi Li, Y uqing Zhao,
Yilei Zhao, Xinyu Cai, et al. 2024a. Finagent: A
multimodal foundation agent for ﬁnancial trading:
Tool-augmented, diversiﬁed, and generalist. arXiv
preprint arXiv:2402.18485.
Y adong Zhang, Shaoguang Mao, Tao Ge, Xun Wang,
Adrian de Wynter, Y an Xia, Wenshan Wu, Ting Song,
Man Lan, and Furu Wei. 2024b. Llm as a master-
mind: A survey of strategic reasoning with large
language models. arXiv preprint arXiv:2404.01230.
Y uzhe Zhang and Hong Zhang. 2023. Finbert–mrc: ﬁ-
nancial named entity recognition using bert under the
machine reading comprehension paradigm. Neural
Processing Letters, 55(6):7393–7413.
Zhihan Zhou, Liqian Ma, and Han Liu. 2021. Trade
the event: Corporate events detection for news-based
event-driven trading. In Findings of the Association
for Computational Linguistics: ACL-IJCNLP 2021 ,
pages 2114–2124, Online. Association for Computa-
tional Linguistics.
2518

Appendices
A Memory Ranking Mechanism of
FINMEM
Upon receiving an investment inquiry, F INMEM re-
trieves the top-K critical memory events from each
layer and channels them to the immediate reﬂection
component of the working memory. These events
are selected based on their information retrieval
score, γE
l , where l represents the layer (shallow,
intermediate, or deep), as deﬁned in Equation 2.
γE
l = SE
Recencyl
+ SE
Relevancyl
+ SE
Importancel
, (2)
where each memory event is only associated
with one score and can only belong to a single layer.
Let E denote a given memory event. The scoring
mechanism for E, adapted from Park et al. ( Park
et al., 2023) but with modiﬁed recency and impor-
tance computations, is tailored to handle data with
various timelines and to achieve layered processing
that represents the diverse periodicities of the ﬁ-
nancial environment. This score encapsulates three
metrics: recency (how recently the event occurred),
relevancy (the event’s pertinence to the current con-
text), and importance (the event’s signiﬁcance). In-
dividual metric scores exceeding 1.0 are scaled to
the [0,1] range before being summed, ensuring a
balanced contribution from each component and
preventing any single metric from dominating the
overall score. The resulting composite score pro-
vides a comprehensive evaluation of the memory
event’s signiﬁcance within the multi-layered, peri-
odically varying ﬁnancial landscape.
SE
Recencyl
= e
− δE
Ql , δ E = tP − tE, (3)
where δE represents the time elapsed be-
tween a memory event’s occurrence and the
trading inquiry’s arrival. The model utilizes
three processing layers, each corresponding to
a speciﬁc timeframe: shallow ( Qshallow = 14
days), intermediate ( Qintermediate = 90 days), and
deep ( Qdeep = 365 days). These intervals rep-
resent two weeks, a quarter, and a year respectively.
When a trade inquiry P arrives in processing
layer l via an LLM prompt, the agent calculates the
recency score SE
Recencyl
for a memory event E using
Equation 3. This score inversely correlates with the
time elapsed between the inquiry and the event’s
memory timestamp, mapping to Ebbinghaus’s for-
getting curve (Murre and Dros, 2015). The stability
term Ql in Equation 3 modulates memory decay
rates across layers, with higher values in deeper
layers indicating longer memory persistence. For
instance, in the trading context, company annual re-
ports (e.g., Form 10-Ks) are assigned higher stabil-
ity values and categorized within deeper processing
layers compared to daily ﬁnancial news, reﬂecting
their extended timeliness, relevance, and impact on
ﬁnancial decision-making.
SE
Relevancyl
= mE · mP
‖mE‖2 × ‖mP‖2
(4)
The relevancy score SE
relevancyl quantiﬁes the se-
mantic similarity between a memory event E and
the current query P using cosine similarity of their
respective embedding vectors, mE and mP, as
shown in Equation 4. These embeddings are gener-
ated from the event’s textual content and the LLM
prompt query (which includes trading inquiries and
the agent’s character setting) using OpenAI’s "text-
embedding-ada-003" model.
The importance score SE
Importancel for a memory
event E in layer l is calculated as the product of a
value vE
l (derived from a uniform piecewise scor-
ing function, Equation 5) and a degrading ratio θl
(Equation 6), as shown in Equation 7. This ap-
proach, adapted from ( Park et al., 2023), is tailored
to our stratiﬁed long-term memory structure. The
likelihood of higher vE
l values increases from shal-
low to deep layers, while θl measures the diminish-
ing importance of an event over time using layer-
speciﬁc exponential functions. The base αl for each
layer follows αshallow < α intermediate < α deep
(set to 0.9, 0.967, and 0.988 respectively), ensuring
θl decreases to a threshold of 5 after 30, 90, and
365 days for shallow, intermediate, and deep lay-
ers. This layered approach, implemented through
three-piece-wise functions for both SE
Importancel and
SRecencylE, enables FinMem to process long-
term memory in a stratiﬁed manner. Memory
events are purged when SRecencylE falls below
0.05 or SImportancel
E is under 5 (pre-scaling),
maintaining the relevance and efﬁciency of the
memory store.
2519

vE
l =



40 with probability p1
60 with probability p2
80 with probability p3
(5)
θl = (αl)δE
, l = shallow, intermediate, deep,
(6)
where p1 + p2 + p3 = 1 , but their
values vary by shallow, intermediate, and
deep processing. when shallow processing
p1, p2, p3 = {0.8, 0.15, 0.05}, intermediate pro-
cessing, p1, p2, p3 = {0.05, 0.8, 0.15} and deep
processing, p1, p2, p3 = {0.05, 0.15, 0.8}.
SE
Importancel
= vE
l ∗ θl, (7)
Furthermore, F INMEM employs an access
counter function to dynamically manage memory
events across layers, ensuring that crucial events
inﬂuencing trading decisions are elevated to deeper
layers for extended retention and recurring access.
This process, monitored by the LLM validation tool
Guardrails AI, tracks critical memory IDs across
layers. Events deemed pivotal for investment suc-
cess receive a 5-point boost to their importance
score ( SE
Importancel). Upon meeting upgrade crite-
ria for a deeper layer, an event’s recency score
(SRecencyl
E) is reset to 1.0, underscoring its sig-
niﬁcance and preventing rapid decay. Conversely,
less relevant events gradually fade. This mecha-
nism allows F INMEM to efﬁciently identify, prior-
itize, and retain key events based on their nature
and retrieval frequency, while gradually phasing
out less impactful information, thereby maintain-
ing a dynamic and relevant memory structure for
ﬁnancial decision-making.
B Details on Evaluation Metrics
Below is a brief overview of these metrics:
Cumulative Return (CR) % measures the total
value change of an investment over time by sum-
ming daily logarithmic returns, shown in Equa-
tion 8. Higher values indicate better strategy effec-
tiveness.
CR =
n∑
t=1
ri =
n∑
t=1
[
ln
( pt+1
pt
)
· actiont
]
(8)
, where ri is the logarithmic return from day t to
t + 1, pt and pt+1 are the closing prices on days t
and t + 1, respectively, and actiont is the model’s
trading decision for day t.
Sharpe Ratio (SR) assesses risk-adjusted returns
by dividing the average excess return ( Rp) over the
risk-free rate ( Rf ) by its volatility ( σp), detailed
in Equation 9. Higher ratios signify better perfor-
mance.
SR = Rp − Rf
σp
(9)
Annualized V olatility (A V) % and Daily V olatil-
ity (DV) % quantify return ﬂuctuations; A V is
derived by scaling DV ( standard deviation of daily
logarithmic returns) by the square root of the an-
nual trading days (252), as in Equation 10. This
metric highlights potential return deviations across
the year.
A V= DV ×
√
252 (10)
Max Drawdown (MDD) % calculates the largest
portfolio value drop from peak to trough, as given
in Equation 11. Lower values indicate lesser risk
and higher strategy robustness.
MDD = max( Ppeak − Ptrough
Ppeak
) (11)
Note that CR and the SR are often considered
more essential than A V and MDD in evaluating
asset trading performance due to their focus on
long-term gains and risk-adjusted returns by their
deﬁnition. Here, we regard these two metrics as
primary metrics when evaluating the experiment
outcomes.
C An example of mixed and lagged
market signals: Partial investment
insights of TSLA on 2021-03-05
Here are some insights from the agent’s memory
module for TSLA as of 2021-03-05. A few memory
records are omitted; these are either neutral or posi-
tive. Despite this, the stock price trend for TSLA is
sharply downward, conﬂicting with the overall pos-
itive ﬁnancial sentiments and market signals. Utiliz-
ing proprietary models such as GPT4 and GPT-o1
as backbones, the ﬁnancial decision-making agent
can leverage other investment insights like histor-
ical momentum and self-reﬂection to consistently
support a ’Sell’ decision. In contrast, the large-
sized open-source models like Qwen2.5-72B and
DeepSeek-67B-Chat exhibit instability in produc-
ing consistent actions across repeated experimental
trials.
2520

Short-term Memory
1 Sentiment: Negative: The key insights from
the news regarding Tesla Inc (NASDAQ:
TSLA) losing market share to Ford Motor
Company’s (NYSE: F) Mustang Mach-E in
the United States are as follows:1. **Mar-
ket Competition**: Tesla is facing increased
competition in the electric vehicle (EV) mar-
ket, particularly from established automakers
like Ford. The Mustang Mach-E’s success in-
dicates that other companies are effectively
entering the EV space, which could impact
Tesla’s dominance.2. **Market Share Im-
pact**: The loss of market share to Ford sug-
gests that Tesla’s growth trajectory in the U.S.
may be challenged. Investors should consider
the potential implications for Tesla’s revenue
and proﬁtability if this trend continues...
2 Sentiment: Positive : The news about a
Banksy artwork being burned and sold as a
non-fungible token (NFT) highlights a few
key insights relevant to investment decisions,
particularly concerning innovative and dis-
ruptive technologies: 1. **Emergence of
NFTs**: The transformation of physical art
into digital assets through NFTs signiﬁes a
growing trend in the digital economy. This
trend could inﬂuence sectors beyond art, in-
cluding technology and ﬁnance, as more in-
dustries explore blockchain applications ...
3 Sentiment: Neutral : To provide a summary
of key insights relevant to making investment
decisions about Tesla (TSLA) from the pre-
open movers news, I would focus on the fol-
lowing aspects: 1. **Stock Performance**:
Look for any signiﬁcant pre-market price
movements for TSLA. If the stock is show-
ing a notable increase or decrease, it could
indicate investor sentiment or reaction to re-
cent news.2. **News Catalysts**: Identify
any speciﬁc news items or announcements re-
lated to Tesla that might be inﬂuencing its
stock price. This could include earnings re-
ports, product launches, regulatory news, or
changes in leadership...
4 Sentiment: positive : The key insights from
the news about Tesla’s vehicle registrations
in Germany are as follows: 1. **Signiﬁcant
Growth in Registrations**: Tesla experienced
a 78% year-over-year increase in vehicle reg-
istrations in Germany as of January. This sub-
stantial growth indicates a strong demand for
Tesla vehicles in one of Europe’s largest au-
tomotive markets.2. **Market Penetration**:
The surge in registrations suggests that Tesla
is successfully penetrating the German mar-
ket, which is traditionally dominated by local
automakers. This could imply a growing ac-
ceptance and preference for electric vehicles
(EVs) in Germany, beneﬁting Tesla as a lead-
ing EV manufacturer...
...
Mid-term Memory
1 Sentiment: Negative : The news about U.S.
Senator Elizabeth Warren’s proposal for a
Democratic wealth tax could have several im-
plications for investment decisions regarding
Tesla (TSLA):1. **Impact on Wealthy In-
vestors**: The proposed wealth tax targets the
wealthiest Americans, which could include
major shareholders and investors in Tesla. If
implemented, this tax could lead to changes in
investment strategies among these individuals,
potentially affecting Tesla’s stock price due
to shifts in large-scale holdings....5. **Long-
term Implications**: While the immediate
impact might be uncertain, the long-term im-
plications of a wealth tax could lead to a more
equitable distribution of wealth, potentially
increasing consumer spending power. This
could beneﬁt Tesla if it results in a larger mar-
ket for its products. Investors in Tesla should
closely monitor the progress of this proposal
and consider its potential effects on market
dynamics and investor behavior.
2 Sentiment: positive : The key insights from
the news information regarding consumer
preferences in China for electric cars, partic-
ularly focusing on price and driving range,
can be relevant for making investment deci-
sions about Tesla (TSLA) in the following
ways: 1. **Market Demand Alignment**:
Tesla’s ability to align its product offerings
with the primary concerns of Chinese con-
sumers 2014price and driving range˘2014could
signiﬁcantly impact its market share and sales
growth in China. Investors should monitor
how Tesla’s models compare to competitors
in these aspects... Overall, Tesla’s success in
2521

addressing the key consumer preferences in
China 2014price and driving range 2014will
be critical for its growth prospects in the re-
gion, making these factors important consid-
erations for investors.
...
Long-term Memory
1 Sentiment: Negative: The key insights from
the news about Tesla Inc. (NASDAQ: TSLA)
facing competition in China are as follows: 1.
**Increased Competition**: Tesla’s Model 3
is experiencing signiﬁcant competition from
a budget electric vehicle produced through
a joint venture involving General Motors
(NYSE: GM). This indicates a more compet-
itive landscape in the Chinese electric vehi-
cle (EV) market, which could impact Tesla’s
market share and sales growth in the region...
For investors, these insights suggest a need
to closely watch Tesla’s strategic responses
to competition in China, its pricing strate-
gies, and any potential impacts on its ﬁnan-
cial performance. Additionally, understanding
the broader competitive landscape and Tesla’s
ability to sustain its growth trajectory will be
crucial for making informed investment deci-
sions.
2 Sentiment: positive : The news about Bill
Gates’ concerns regarding Bitcoin’s impact
on climate change highlights a broader issue
of environmental sustainability in the tech
and ﬁnancial sectors. Here are the key in-
sights relevant to making investment decisions
about Tesla (TSLA): 1. **Environmental Im-
pact Awareness**: Bill Gates’ concerns un-
derscore the growing awareness and scrutiny
of the environmental impact of technology
and ﬁnancial products. This is relevant for
Tesla, as the company positions itself as a
leader in sustainable energy and electric ve-
hicles (EVs)... Overall, the emphasis on en-
vironmental impact and sustainability in the
tech sector could reinforce Tesla’s strategic ad-
vantages and appeal to investors prioritizing
green investments.
...
D Extended Task: Multi-Asset Portfolio
Management Task
Besides the single-asset trading tasks presented
in the paper, I NVESTOR BENCH is able to incor-
porate the multi-asset portfolio management task,
which further illustrates I NVESTOR BENCH s gener-
alization capabilities. This advanced task involves
more sophisticated trading strategies and mathemat-
ical inference, allowing the agent to dynamically
allocate asset weights at each decision step and
rebalance the portfolio based on current market
conditions.
The experimental results for a compact portfo-
lio consisting of TSLA, JNJ, and UVV are shown
in Table 5, using three LLMs as agent backbones
(one from each size category). The I NVESTOR -
BENCH agent’s performance on this portfolio man-
agement task across LLMs aligns with the overall
trend with the single-asset trading task presented in
Figure 3b in our paper, the agent with the large-size
LLM delivers the best performance, while that with
small-size and mid-size LLMs performs lower and
closely in terms of primary evaluation metrics, Cu-
mulative Returns (CR) and Sharpe Ratio (SR). This
illustrates the consistency of our decision-making
agent performance across distinct types of LLMs
for various task types and trading strategies.
Table 5: Comparison of performance across LLMs of
varying sizes in multi-asset trading tasks
Model Type CR SR A V MDD
Llama-3.1-70B-Instruct 31.473 1.482 17.100 11.205
Yi-1.5-34B-Chat 15.458 0.903 13.769 9.883
Llama-3.1-8B-Instruct 20.255 0.953 17.106 11.219
E Case Study on I NVESTOR BENCH in
Extreme Market Conditions
Table 6: Comparison of performance across LLMs of
varying sizes in the extreme market conditions
Model Type CR SR A V MDD
Buy & Hold -56.738 -0.936 53.911 52.078
GPT-4 10.163 0.433 20.866 12.672
Llama-3.1-70B-Instruct -27.596 -0.633 38.708 34.797
Yi-1.5-34B-Chat -32.723 -0.606 48.012 41.197
Llama-3.1-8B-Instruct -37.56 -0.736 45.350 47.748
Here, we use a case study to demonstrate I N-
VESTOR BENCH ’s robust performance during peri-
ods of extreme market featured by signiﬁcant mar-
ket volatility.
2522

Table 7: Performance Comparison of LLMs in Multi-Asset Trading Tasks under Different Market Conditions
Model Type Bullish Bearish Mixed Signal
CR SR A V MDD CR SR A V MDD CR SR A V MDD
Buy and hold 51.987 3.654 36.322 12.551 -96.218 -2.884 60.645 65.24 39.244 0.600 52.339 37.975
GPT-4 27.219 2.095 33.163 12.551 61.236 1.781 62.513 18.759 45.246 0.821 44.088 25.031
Llama-3.1-70B-Instruct 43.926 3.144 36.349 12.551 20.819 0595 63.612 40.686 37.545 0.615 48.862 29.813
Yi-1.5-34B-Chat 17.571 1.790 25.062 5.617 10.424 0.317 59.668 41.988 35.364 0.558 50.757 35.49
Llama-3.1-8B-Instruct 18.685 1.417 33.655 9.778 -39.44 -1.386 51.737 44.274 35.622 0.574 49.636 36.383
We evaluated the asset, TSLA, by examining Cu-
mulative Returns (CRs) and Sharpe Ratios (SRs),
Annualized V olatility (A V) and Maximum Draw-
down(MDD). Our dataset spans a training period
from 2022-01-17 to 2022-03-31 and a test period
from 2022-04-01 to 2022-10-15. These periods
are selected because the VIX (CBOE V olatility
Index) remained elevatedaveraging above 20indi-
cating heightened market volatility.
We present the I NVESTOR BENCH agent’s per-
formance using four representative LLMs as back-
bonesone proprietary model (GPT-4) and three
open-source models of varying sizes (Llama-3.1-
70B-Instruct, Yi-1.5-34B-Chat, and Llama-3.1-8B-
Instruct)chosen for their relatively stable perfor-
mance (See Table 6 below). Overall, the agents
performance with each LLM aligns with the re-
sults in Table 2 as well as Figure 3a and 3b in
our paper. Speciﬁcally, 1) The proprietary LLM
outperforms the open-source models. 2) Larger
model parameter sizes consistently lead to higher
decision-making quality and robustness.
F Case Study on I NVESTOR BENCH
across Different Market Scenarios
We introduced an additional set of case studies
in bullish, bearish, and mixed-signal market envi-
ronments to highlight the robustness of the agents
performance in I NVESTOR BENCH .
We assess the I NVESTORBENCH agent’s perfor-
mance robustness using three separate time periods
with distinct market trends on the same assets, task
type, and LLMs. For the bullish market, the train-
ing period spanned from 2023-03-06 to 2023-05-
01, followed by a test period from 2023-05-02 to
2023-07-10. For the bearish market, the training
period ran from 2022-08-08 to 2022-09-26, and the
test period from 2022-09-27 to 2023-01-02. We
present Bullish and Bearish results below Table 7.
For the mixed market condition, we employed the
same dates speciﬁed in Section 4.1 of the paper.
In the mixed-signal and bearish market condi-
tions, the agents performance trends across various
LLMs remain closely aligned with the TSLA trad-
ing case shown in Table 2 as well as Figure 3a and
3b of our paper. In the bullish market scenario,
while larger LLMs generally maintain stronger per-
formance, the simple Buy-and-Hold strategysup-
ported by a continuously rising asset priceoutper-
forms strategies provided by the agent that con-
sider multiple investment perspectives. The latter
approach can be inﬂuenced by noisy market sig-
nals, such as occasional neutral or negative news.
The proprietary LLM GPT-4, characterized by its
comprehensive reasoning, exhibits more caution
and thus lower returns compared to the large-scale
open-source model Llama-3.1-70B-Instruct. How-
ever, both still surpass other LLMs, consistent with
several asset trading cases reported in Table 2.
Our case study proves that the I NVEST BENCH
agent’s performance stays robust and consistent
with the analysis outcomes in our paper in general.
G Case Study on Single Stock Trading:
Forecast for TSLA on 2022-10-25 to
Predict Trading Decision on 2022-10-26
2523

Initialize Proﬁle
1. Operations:
- Provide a performance overview of the trading stock based on available data.
- Set up the risk inclination as the key character of the trading agent.
2. Range: Financial information such as the ﬁnancial sectors, historical performance, and previous stock trends of the
trading stock.
3. Prompts: Y ou are an experienced trading manager and investment ﬁrm. Y our task is to make informed decisions on
the given stock based on the provided information.
Under Self-Adaptive Risk Character Setting: When historical momentum is positive, you are a risk-seeking investor.
But when historical momentum is negative, you are a risk-averse investor.
4. General background setting:
Y ou have accumulated a lot of information about the following sectors, so you are especially good at trading them:
1)Electric V ehicles (Automotive Sector). 2) Energy Generation and Storage...From year 2021 to 2022 September,
Tesla’s continued growth and solid ﬁnancial performance over the deﬁned period ...
Summarize
1. Operations:
- Summarize different types of input information.
- Distribute them to corresponding layers of the long-term memory database.
2. Range: Daily market news, Long Documents such as company 10-K and 10-Q reports
3. Prompts:
- (1). Summarize the contents: Summarize the following documents into 1000 words.
- (2). Comprehend the investment sentiment of news insights: The positive, neutral and negative scores are for
understanding the investment sentiments, opinions, or emotions. For example, positive news about a company can lift
investor sentiment, encouraging more buying activity, which in turn can push stock prices higher...
4. Outputs:
(1). To Shallow Memory Layer:
- [News (ID: 261)] Here’s How Much Y ou Would Have Made Owning Tesla Stock In The Last 10 Y ears Tesla
(NASDAQ:TSLA) has outperformed the market over the past 10 years by 50.69% on an annualized basis producing an
average annual return of 60.76%. Currently, Tesla has a market capitalization of $683.54 billion.... The sentiment is
{positive}.
- [News (ID: 278)] Tesla Q3 Earnings Are Imminent. Can Nio Foreshadow What’s To Come? What To Know Before
The Print Tesla Inc (NASDAQ: TSLA) shares were trading down slightly Wednesday afternoon ahead of the automaker ´s
third-quarter report, but the stock is up 6% over the last ﬁve sessions... The sentiment is {positive}.
- ...
(2). To Intermediate Memory Layer:
- [Form 10-Q (ID: 222)] Tesla Q3 2022 revenues were $21.5 billion, up 56% year-over-year. Automotive sales revenue
grew 56% to $17.8 billion driven by higher Model 3/Y and Model S/X deliveries. Gross automotive margin declined to
27.9% due to cost inﬂation and factory ramps. Net income was $3.3 billion, up 102% year-over-year. Positive free cash
ﬂow was $6.1 billion...
- [News (ID: 275)] Tesla Q3 Earnings Highlights: Record Revenue, Operating Margin And Free Cash Flow, Tesla Semi
Deliveries Coming In December Electric vehicle leader Tesla Inc (NASDAQ: TSLA) reported third-quarter ﬁnancial
results after market close Wednesday...The sentiment is {neutral}.
- [News (ID: 274)] Tesla Preps For 2023 Cybertruck Launch, Will Make Battery Packs In California The Cybertruck is
one of Tesla Inc. (NASDAQ: TSLA) most hotly anticipated, but also most delayed, products. - ...The sentiment is
{negative}.
(3). To Deep Memory Layer:
- [News (ID: 161)] Tesla Whale Trades Spotted A whale with a lot of money to spend has taken a noticeably bearish
stance on Tesla. Looking at the options history for Tesla (NASDAQ:TSLA) we detected 477 strange trades. The
sentiment is {positive}.
- [Self-reﬂection (ID: 226)] Given the short-term positive news score in the market for TSLA and a positive cumulative
return, there is a high probability of continued growth in the short term. However, investor should be aware of potential
threats in the mid-term market with competitors like General Motors, and Nio...
2524

Observe
1. Operations: Access and interpret market indicators such as current stock prices and historical momentum data.
2. Range: Stock’s daily adjusted closing price, historical momentum in the past k days (k = 3 in this case), etc.
3. Prompts:
- The information below provides a summary of stock price ﬂuctuations over the previous few days, which is the
"momentum" of a stock. It reﬂects the trend of a stock. Momentum is based on the idea that securities that have
performed well in the past will continue to perform well, and conversely, securities that have performed poorly will
continue to perform poorly.
4. Outputs:
- (1). The daily adjusted closing price of TSLA on {2022-10-25} is {$222.42}.
- (2). Train: On {2022-10-25}, the momentum of TSLA, indicated by the price difference between the current and the
next trading
day, is {$2.22}.
Test: On {2022-10-25}, the historical momentum of TSLA, as measured by its cumulative logarithmic returns in
the past
{3} days, was {7.05%}.
Figure 4: First section of INVESTORBENCH’s workﬂow for perceiving and processing multi-sourced information from
market environment.
Reﬂect
1. Operations:
Train: - Infer the reasoning from the retrieved memories insights for the recommended trading actions the increase or
decrease of the trading stock price.
Test: - Generate recommended trading actions. - Infer the reasoning for making such trading action.
Train & Test: - Track the memory insight IDs that contributed to the current trading decision.
2. Range: Interpretation of observed market indicators and top k insights from each layer of the long-term memory
database.
3. Prompts:
Train: Given the following information, can you explain why the ﬁnancial market ﬂuctuation from the current day to the
next day behaves like this? Summarize the reason for the decision. Y ou should provide summary information and the
IDs of the information to support your summary.
T est:Given the information, can you make an investment decision? Just summarize the reason for the decision. Consider
only the available short-term information, the mid-term information, the long-term information, and the reﬂection-term
information. Consider the momentum of the historical stock price. Also, consider how much share of the stock the
investor holds now. Y ou should provide exactly one of the following investment decisions: buy or sell. Y ou could go
with the ’hold’ option if necessary. Y ou also need to provide the IDs of the information to support your decision.
4. Outputs:
Train:
- Recommended Action Reason: The recommended trading action is ’Buy’. This can be attributed to positive market
sentiment driven by a range of favorable factors. Notably, Tesla’s strong battery supply chain, effective market
expansion plans in Germany, and its ability to maintain market strength despite potential competitive threats positively
inﬂuenced investor conﬁdence, contributing to the stock price rise.
- Reﬂected Contributed Insights IDs: {261, 278, 275, 274, 161,...}
Test:
- Trading Action: {Buy}
- Recommended Action Reason: Considering the short-term positive revenue growth, higher Model 3/Y and Model S/X
deliveries, positive free cash ﬂow, and increased cash and cash equivalents, Tesla appears to have strong ﬁnancial health.
The momentum being positive, it also encourages a risk-seeking strategy. The signiﬁcant positive signs from news such
as the imminent Q3 earnings add to the favorable view of the investment. Despite the mid-term neutral sentiment,
long-term information shows only marginal drops and the stock is still looking strong overall. These, along with positive
reﬂection-term information, form the reasoning behind the ’buy’ decision...
- Reﬂected Contributed Insights IDs: {261, 278, 275, 274, 161,...}
Figure 5: Second section of INVESTORBENCH’s workﬂow for generating trading action, reasoning and reﬂection.
2525