Available via license: CC BY
Content may be subject to copyright.
ORIGINAL RESEARCH
published: 18 May 2022
doi: 10.3389/frai.2022.865950
Frontiers in Artificial Intelligence | www.frontiersin.org 1May 2022 | Volume 5 | Article 865950
Edited by:
Ronald Hochreiter,
Vienna University of Economics and
Business, Austria
Reviewed by:
Bowei Chen,
University of Glasgow,
United Kingdom
Bertrand Kian Hassani,
University College London,
United Kingdom
*Correspondence:
Rei Taguchi
s5abadiee@g.ecc.u-tokyo.ac.jp
Specialty section:
This article was submitted to
Artificial Intelligence in Finance,
a section of the journal
Frontiers in Artificial Intelligence
Received: 30 January 2022
Accepted: 06 April 2022
Published: 18 May 2022
Citation:
Taguchi R, Watanabe H, Sakaji H,
Izumi K and Hiramatsu K (2022)
Constructing Equity Investment
Strategies Using Analyst Reports and
Regime Switching Models.
Front. Artif. Intell. 5:865950.
doi: 10.3389/frai.2022.865950
Constructing Equity Investment
Strategies Using Analyst Reports and
Regime Switching Models
Rei Taguchi 1
*, Hikaru Watanabe 2, Hiroki Sakaji 1, Kiyoshi Izumi 1and Kenji Hiramatsu 3
1School of Engineering, The University of Tokyo, Tokyo, Japan, 2Faculty of Engineering, The University of Tokyo, Tokyo,
Japan, 3IFIS Japan Limited, Tokyo, Japan
This study demonstrates whether analysts’ sentiments toward individual stocks are
useful for stock investment strategies. This is achieved by using natural language
processing to create a polarity index from textual information in analyst reports. In
this study, we performed time series forecasting for the created polarity index using
deep learning, and clustered the forecasted values by volatility using a regime switching
model. In addition, we constructed a portfolio from stock data and rebalanced it at each
change point of the regime. Consequently, the investment strategy proposed in this study
outperforms the benchmark portfolio in terms of returns. This suggests that the polarity
index is useful for constructing stock investment strategies.
Keywords: BERT, Hidden Markov Model, trading strategy, financial market, regime switching model
1. INTRODUCTION
In recent years, the explosive development of artificial intelligence has brought worldwide attention
to the use of alternative data, which is particularly prominent in the financial and economic fields
and is beginning to be widely used for economic forecasting and investment strategies together with
traditional data. In Japan, the Japan Association for the Promotion of Alternative Data (JADAA)1
and other cross-industry organizations have been established to promote its use actively.
This study is expected to expand the application possibilities of alternative data. Additionally,
this study aims to encourage the expansion of artificial intelligence techniques in investment
technology by providing the same reliability criteria as existing economic statistics in this area
using indexes produced from alternative data. Among them, text data is particularly versatile and
can extract predictions and classifications based on the writing style of the author and the content
bais (Izumi and Sakaji, 2019; Sakaji et al., 2019).
Moreover, this study is expected to expand the possibilities of data linkage. The purpose of
data linkage in this study is to encourage the expansion of artificial intelligence technology in the
investment field by providing the same reliability criteria as existing statistical information in the
investment field through alternative data analysis.
In addition, analyst reports are considered to have a very high information value among text data
(Hiramatsu et al., 2021). Analyst reports contain rating information on individual stocks; therefore,
the polarity index generated from these reports is assumed to affect stock prices, Return on Equity
and Price Earnings Ratio. The polarity index is a time series plot of the polarity assigned to the text
data. In recent years, it has been found that polarity index generated from analyst reports are ahead
1https://alternativedata.or.jp
Taguchi et al. Constructing Equity Investment Strategies
of macroeconomic indexes such as exchange rates and
government bonds issuance (Taguchi et al., 2021) and are
expected to be applied to investment technology.
Although there are many studies on portfolio management
methods using machine learning (Wang et al., 2020; Yun et al.,
2020; Zhang et al., 2020; Chen et al., 2021; Ma et al., 2021), this
study is novel as we utilize the method of sentiment analysis into
portfolio management.
This study shows that sentiment toward the future of
individual stocks in analyst reports is useful in investment. The
expected outcome is that the investment strategy proposed in this
study will generate returns exceeding the benchmark portfolio.
Consequently, it is envisioned that changes in the regime of polar
indicators generated from analyst reports will serve as signals to
change the investment allocation of individual stocks.
This study’s contribution is to demonstrate that an investment
strategy can generate returns by using analysts’ sentiments
toward individual stocks to signal portfolio asset allocation.
From this, we demonstrate that analyst reports are useful
for investment.
2. RELATED WORKS
There is considerable literature on portfolio management
methods using machine learning and deep learning. Ma et al.
(2021) performed return prediction with several opportunity
learning models, including deep multilayer perceptron, and
incorporated the prediction results into the advanced mean-
variance model for portfolio optimization. Yun et al. (2020)
proposed a two-stage deep learning model for a forecast-based
Exchange Traded Fund (ETF) portfolio management approach.
Wang et al. (2020) proposed a portfolio management method
that considers the long-term dependence of time series variability
using mean-variance models and deep learning theories such as
Long Short Term Memory (LSTM). Zhang et al. (2020) proposed
a deep learning framework for maximizing the Sharpe ratio of a
portfolio. Chen et al. (2021) proposed a new model to optimize
XGBoost with an improved firefly algorithm (IFA) and perform
portfolio optimization.
Some studies have been conducted on creating polarity
indicators by machine learning. In Yono et al. (2020), Financial
and Economic Statistics Monthly were trained using LSTM, and
indicators for each topic such as consumer spending, capital
investment, and inventory were created using Latent Dirichlet
Allocation (LDA). They also analyzed which macro factors
are more influential by determining how much each topic’s
sentiment contributes to the overall sentiment. Katayama et al.
(2019) developed a sentiment polarity identification model for
finance by processing the Japan Economic Watcher Survey with
Word2vec and training it with LSTM.
The above studies use texts with similar contents, such as the
Financial and Economic Statistics Monthly and the Economic
Watcher Survey. In contrast, our study differs in that we use
analyst reports as an indicator by summing up the sentiment
of individual stocks. Some studies have been conducted as an
example of research on text mining using analyst reports. In
Hiramatsu et al. (2021), stock prices respond strongly to the
sentiment of the report, and drift in stock prices is observed after
the report is issued, suggesting that the textual information in
analyst reports is useful in asset management practice. Further,
Asquith et al. (2005) investigates the association between market
returns and analyst reports content using regression analysis. In
addition, Suzuki et al. (2020) opinion and non-opinion text are
extracted from analyst reports using LSTM and other methods,
and forecasts of net income and stock prices are made.
Our study’s use of Bidirectional Encoder Representations from
Transformers (BERT) to train analyst reports is also different
from other studies. BERT is a Transformer-based language
model. An example of a study using BERT is Hiew et al. (2019)
where text data is given polarity by BERT using Weibo, a Chinese
Social Networking Service, and stock price predictions are made
using LSTM.
Furthermore, our study differs from other studies in that we
use Bi-directional Gated Recurrent Unit (BiGRU) and Hidden
Markov Model with Gaussian Mixture Model emissions (GMM-
HMM) to trigger the portfolio rebalancing. An example of a study
using BiGRU in the financial field is Chen et al. (2020) which used
labeled financial tweet data and BiGRU to perform sentiment
analysis. Liu et al. (2021) used GMM-HMM and LSTM to predict
stock prices. BiGRU and LSTM have several applications outside
the financial sector (Talha et al., 2017; Yan et al., 2021).
3. METHODS
In this study, we performed asset allocation for a stock portfolio
composed of two stocks, using signals as change points in the
polar indicators regime created from analyst reports. We used
analyst reports to develop an investment strategy using natural
language processing and artificial intelligence techniques in the
following four steps. We demonstrate that analyst reports are
useful for investment by comparing this investment strategy with
the benchmark strategy.
Step 1: BERT is used to learn the analyst report text and
classify the polarity into three values: “positive,” “negative,” and
“neutral.” A detailed description of the method is given in section
3.1, and the results are shown in section 5. This step aims to create
a polarity index.
Step 2: We perform time series forecasting for the created
polarity index (by industry) using a neural network. The
algorithm used is Bi-directional Gated Recurrent Unit (BiGRU).
A detailed description of the method is given in section 3.2, and
the results are shown in section 5. This step in forecasting polar
indicators aims to make a preliminary step in creating signals for
future asset allocations.
Step 3: A three-month moving average is taken for the
forecast value calculated in Step 2, and the volatility is clustered
into “high,” “medium,” and “low” states using the regime-
switching model. The algorithm used is the Hidden Markov
Model with GMM-HMM. A detailed description of the method is
provided in section 3.3 and the results are presented in section 5.
This step creates a signal for asset allocation at a future point that
Frontiers in Artificial Intelligence | www.frontiersin.org 2May 2022 | Volume 5 | Article 865950
Taguchi et al. Constructing Equity Investment Strategies
FIGURE 1 | Architecture.
has not yet manifested by dividing the polarity index predicted in
Step 2 into three regimes.
Step 4: We retrieved monthly stock price data from Yahoo!
Finance2. We selected two stocks for each of the 33 industries and
created a portfolio for each industry using the model selected in
Steps 2 and 3. Then we rebalanced the portfolio at the regime’s
turn assigned in Step 3 above. In addition, we compare this
strategy with the benchmark strategy and tabulate the results. A
detailed description of the method is given in section 3.4, and the
results are shown in section 5.
The architecture used in this study can be described as shown
in Figure 1.
We created a polarity index as discussed. The assumption is
that the sentiment of individual stocks is included in this polarity
index, which is used for the investment experiment.
3.1. Bidirectional Encoder Representations
From Transformers (BERT)
BERT is a Transformer-based language model proposed by
Devlin et al. (2019). The reason we chose BERT for this study
is because it has demonstrated its effectiveness against other
models in eight benchmarks in Devlin et al. (2019). BERT enables
successful bidirectional learning by masking some tokens in the
input and predicting the masked words. This study used the
model3published by Inui and Suzuki Laboratory at Tohoku
University as a trained model for Japanese BERT. Additionally,
we use the method of Taguchi et al. (2021) to create two types
of models (UP Model and DOWN Model) using BERT. In the
Taguchi et al. (2021) method, a report with a change in rating
to “buy” is considered to be 1, and 0 otherwise, and a report
with a change in rating to “sell” is considered to be −1. The final
2https://finance.yahoo.co.jp/
3https://github.com/cl-tohoku/bert- japanese
TABLE 1 | Polarity judgment method.
UP model DOWN model Final judgment
0 (no change) −1 (sell) −1 (sell)
1 (buy) 0 (no change) 1 (buy)
1 (buy) −1 (sell) High likelihood
0 (no change) 0 (no change) 0 (no change)
judgment method is shown in Table 1. Details of this method can
be obtained by referring to Taguchi et al. (2021). In addition, this
architecture replaces the model in Sakaji et al. (2008) with BERT.
See section 5 for the calculation results.
3.2. Time Series Forecasting Using
Bi-Directional Gated Recurrent Unit
(BiGRU)
Section 3.2 aims to take a preliminary step in creating a
signal for future asset allocation. Time series forecasting creates
pseudo-sentiment information not apparent. Gated Recurrent
Unit (GRU) is a neural network consisting of Reset and Update
gates, proposed by Cho et al. (2014). GRU is a neural network
that can deal with the gradient vanishing problem of Recurrent
Neural Network (RNN) and has a low computational cost.
This point was corroborated by the verification of performance
in section 5.
The architecture of GRU is represented by Figure 2.
BiGRU is a model that can add past and future information
to GRU.
The architecture of BiGRU is represented by Figure 3. Details
of the GRU and BiGRU algorithms are given in Appendix A.
Time series forecasting was evaluated using Root Mean
Squared Error (RMSE). RMSE is a measure to evaluate the
Frontiers in Artificial Intelligence | www.frontiersin.org 3May 2022 | Volume 5 | Article 865950
Taguchi et al. Constructing Equity Investment Strategies
model’s error, and the smaller it is, the better the model’s
performance. The equation for RMSE is as follows.
RMSE =q1
nPn
i=1yi− ˆyi2(1)
RMSE is calculated by dividing the observed values by yi(i=
1, 2, 3, ...,n), and the calculated value (predicted value) from the
model is ˆyi.
FIGURE 2 | Architecture of GRU.
3.3. Regime Assignment Using Hidden
Markov Model With Gaussian Mixture
Model Emissions (GMM-HMM)
In section 3.3, we divide the polarity index predicted in section
3.2 into three regimes to create a signal for asset allocation in
the future not yet manifested. The changes in the regimes are
assumed changes in the environment and are used as signals.
This study used the Hidden Markov Model (HMM) with the
Gaussian Mixture Model (GMM) state output distribution to
assign regimes to the predicted values of the polarity index. GMM
is a statistical model that can model N populations following
a normal distribution, while HMM is a Markov model based
on a Markov process with hidden states. The architecture of
GMM-HMM is represented by Figure 4.
Details of the GMM-HMM algorithms are given in
Appendix B.
3.4. Investment Simulation
In this section, we design an investment simulation using the
models in sections 3.2 and 3.3. As a proposed method, when the
volatility of the polarity index predicted by the GMM-HMM with
three regimes is “low,” we increase the allocation to stocks with
high monthly volatility. In addition, when the predicted volatility
of the polarity index is “medium,” the investment allocation is
set at 50%. Meanwhile, if the forecasted volatility of the polarity
index is “high,” we increase the allocation to stocks with low
monthly volatility. The monthly volatility of a stock is defined
as the maximum value among the current month’s True Range4
÷Typical Price for the current month5×100, which is calculated
below. We assume that the above portfolio rebalancing strategy
generates higher returns than the benchmark strategy.
4Defined in this article as high-low, high-close, and close-low.
5Average of the high, low, and close prices.
FIGURE 3 | Architecture of BiGRU.
Frontiers in Artificial Intelligence | www.frontiersin.org 4May 2022 | Volume 5 | Article 865950
Taguchi et al. Constructing Equity Investment Strategies
FIGURE 4 | Architecture of GMM-HMM.
TABLE 2 | Notation of investment simulations.
Notation
RoPI Regime of polarity index (volatility low =0, medium =1, high =2)
Am Amount of stock purchased
NoR Number of rebalancing (1,...,n)
rInvestment ratio (0 ≤r≤1)
BLBuying price for stocks with “low” volatility
SLSelling price for stocks with “low” volatility
BHBuying price for stocks with “high” volatility
SHSelling price for stocks with “high” volatility
BαBuying price at the begin of the investment simulation for α
SαSelling price at the end of the investment simulation for α
BβBuying price at the begin of the investment simulation for β
SβSelling price at the end of the investment simulation for β
See Table 2 for Notation in the investment simulation. The
total profit and loss of the portfolio Pare defined below. We
assume that the portfolio consists of two stocks types.
P=PL+PN+PH(2)
PLis the sum of the portfolio’s profit and loss when the polarity
index volatility is “low,” PNis the sum of the portfolio’s profit
and loss when the polarity index volatility is “medium,” and PH
is the sum of the portfolio’s profit and loss when the polarity
index volatility is “high.” The total of each profit and loss can be
formulated as follows.
PL=Am ∗PNoR
n=1{(1 −rn)∗(SL
n−BL
n)+rn∗(SH
n−BH
n)},
if RoPI is 0 (3)
PN=Am ∗PNoR
n=1{0.5 ∗(SL
n−BL
n)+0.5 ∗(SH
n−BH
n)},
if RoPI is 1 (4)
PH=Am ∗PNoR
n=1{rn∗(SL
n−BL
n)+(1 −rn)∗(SH
n−BH
n)},
if RoPI is 2 (5)
Conversely, the total profit and loss of the benchmark strategy
to be compared, PBM, is the profit and loss of the two stocks
purchased at the beginning of the experiment with the investment
ratio of r=0.5 and sold at the end of the experiment period. PBM
is formulated as follows:
PBM =Am ∗ {0.5 ∗(Sα−Bα)+0.5 ∗(Sβ−Bβ)}(6)
In comparing the benchmark and proposed strategies, we assume
that both portfolios are composed of two similar stocks.
The overall picture of the investment simulation is shown in
Figure 5. In Figure 5, we set the investment ratio r=0.9. For the
results of the investment simulation, see section 5.
Frontiers in Artificial Intelligence | www.frontiersin.org 5May 2022 | Volume 5 | Article 865950
Taguchi et al. Constructing Equity Investment Strategies
FIGURE 5 | Example of investment simulations.
FIGURE 6 | All-industry polarity index with 3-month moving average.
4. EXPERIMENTS
Analyst report data (373,050) was used to create the BERT
polarity index. Utilizing the technologies of IFIS Japan Ltd,
charts and unnecessary texts in research reports were removed.
The data period is from Jan 2010 to December 2020. Of the
data used in BERT, the train, valid, and test ratios are 7:1:2,
respectively. In this study, five models were used for time series
forecasting. The first is Bi-directional GRU (BiGRU); the second
is GRU; the third is Bi-directional LSTM (BiLSTM); the fourth
is Recurrent Neural Network (RNN), and the fifth is LSTM. We
take three-month moving averages for the all-industry polarity
Frontiers in Artificial Intelligence | www.frontiersin.org 6May 2022 | Volume 5 | Article 865950
Taguchi et al. Constructing Equity Investment Strategies
index in preprocessing the data. We use monthly stock price data
obtained from Yahoo! Finance for the investment simulation. See
Appendix C for the stocks used in the experiments.
5. RESULTS
We created a polarity index for all industries and each industry.
The all-industry polarity index was used for model selection, and
the industry-specific polarity index for investment simulation.
The all-industry polarity index is the sum of all outputs (1, 0, −1)
of the industry-specific polarity index.
The 3-month moving averages of the all-industry polarity
index, sorted by time series, are shown in Figure 6.
The results of the comparison experiments are shown in
Table 3. We chose BiGRU, the smallest RMSE among the five in
this study.
TABLE 3 | Time series prediction.
Model RMSE
BiGRU 0.1376
GRU 0.1600
BiLSTM 0.2011
RNN 0.2115
LSTM 0.2156
TABLE 4 | Assigning regimes.
Model Log likelihood
GMM-HMM −72.3536
Gaussian-HMM −72.3544
MS-DRM −84.0420
In this study, we used three models for clustering. The first
is GMM-HMM, the second is Gaussian Hidden Markov Model
(Gaussian-HMM), and the third is Markov witching Dynamic
Regression Models (MS-DRM). In pre-processing the data, three-
month moving averages are taken for the predicted values of the
all-industry polarity index. The log-likelihood was used as the
evaluation index; the larger the model’s log-likelihood, the better
the model’s fit.
The results of the comparison experiments are shown in Table 4.
We chose the GMM-HMM with the largest log-likelihood among
the three in this study.
The results of assigning regimes to the predictions of the
polarity indexes using GMM-HMM are shown in Figure 7.
The investment simulation results in section 3.4 are shown.
Monthly stock price data is obtained from Yahoo! Finance.
For details of the investment simulation, see section 3.4. This
experiment assumes that the amount of stocks Am =12, 000.
See Appendix C for the constituent stocks of each portfolio.
The investment simulation results are shown in Table 5. In the
table, ◦refers to the victory of the proposed strategy against the
benchmark, and •refers to the defeat of the proposed strategy
against the benchmark. The NoR in the table refers to the number
of times the 33-industry portfolio is rebalanced.
6. DISCUSSION
First, Table 5 reveals that the 15
33 industry wins when the
investment ratio is r=0.8, and the 16
33 industry wins when
r=0.9. This indicates that the strategy of increasing the
allocation to stocks with high monthly stock volatility when the
predicted values volatility of the polarity index is “low,” setting the
investment allocation to 50% when the predicted values volatility
of the polarity index is “medium,” increasing the allocation to
stocks with low monthly stock volatility when the predicted
values volatility of polarity index is “high,” is somewhat effective
against the benchmark strategy.
FIGURE 7 | Predicted values for all-industry polarity index with regimes.
Frontiers in Artificial Intelligence | www.frontiersin.org 7May 2022 | Volume 5 | Article 865950
Taguchi et al. Constructing Equity Investment Strategies
TABLE 5 | Winners and losers of proposed strategies in benchmarking strategies.
Investment ratio (r)
Industry name NoR r= 0 r= 1 r= 2 r= 3 r= 4 r= 5 r= 6 r= 7 r= 8 r= 9 r= 10
Fisheries, Agriculture and Forestry 10 •••◦◦◦◦◦◦◦ ◦
Mining 5 •••◦◦◦◦◦◦◦ ◦
Construction 5 ••••••◦◦◦◦ ◦
Foodstuffs 8 ◦◦◦◦◦◦◦◦◦◦ ◦
Textiles 7 •••••••••• •
Pulp and Paper 5 ◦◦◦◦◦◦◦◦◦◦ ◦
Chemistry 5 ••••◦◦◦◦◦◦ ◦
Pharmaceuticals 4 ••••••••◦◦ ◦
Oil and Coal Products 3 •••••••••• •
Rubber Products 6 ◦◦◦◦◦◦◦◦◦◦ ◦
Glass, Clay and Stone Products 5 ◦◦◦◦◦••••• •
Steel 5 ◦◦◦◦◦◦•••• •
Nonferrous Metals 4 •••••••••• •
Metal Products 5 ••••••••◦◦ ◦
Machinery 7 ◦◦◦◦◦◦◦◦◦◦ ◦
Electronic Equipment 3 ◦◦◦◦◦••••• •
Transport Equipment 7 •••••••••• •
Precision Equipment 4 •••••••••• •
Other Products 8 •••••••◦◦◦ ◦
Electricity and Gas 4 ◦••••••••• •
Land Transportation 3 ••••••◦◦◦◦ ◦
Marine Transportation 6 •••••••••• •
Air Transportation 5 •••••••••• •
Warehousing and Transportation 5 ◦••••••••• •
Information and Communication 4 ••••◦◦◦◦◦◦ ◦
Wholesale business 6 ••••••◦◦◦◦ ◦
Retailing 6 •••••••••• •
Banking 7 •••••••••◦ ◦
Securities and Commodity Futures 10 •••••••••• •
Insurance 5 ◦◦◦◦◦◦◦◦◦◦ •
Other Financial Services 5 •••••••••• •
Real Estate 7 •••••••••• •
Service 7 ◦◦◦◦•••••• •
Second, among the portfolios of 33 industries, we observed
seven industries in which the profit and loss of the benchmark
strategy were negative. These include pulp and paper, glass and
stone products, and electric equipment. For these industries,
even when the investment ratio rwas changed, many of them
showed extreme trends such as winning all patterns and losing
all patterns. In some cases, such as glass and stone products
and electrical equipment, the results contradict our hypothesis.
Although it is impossible to determine whether the tendency is
industry-specific or market-influenced only from the results of
this experiment, we would like to explore a method to determine
the investment ratio based on these results when constructing a
portfolio among industries.
Third, we found that the number of NoR varies significantly
among industries. For example, the number of NoR in the
fisheries, agriculture, and forestry industry is ten, whereas the
land transportation industry is three. This indicates that the
tendency to change the volatility of polar indexes is remarkably
different among industries.
Fourth, in Table 3, we expected BiLSTM to have the lowest
RMSE because its mechanism is richer among the five; however,
BiGRU has the lowest RMSE. Chung et al. (2014) shows that
the superiority or inferiority of LSTM and GRU cannot be
determined in general but depends on the data set and the given
task. The same can be said for BiLSTM, LSTM, BiGRU, and GRU
in this experiment.
Fifth, there is still room for improvement in the duration of
signal creation, as shown in the GMM-HMM results (Table 4).
Since, in practice, the asset allocation of portfolios is often
done monthly, we used monthly data for both text and
stocks. It is necessary to use frequent data to create more
detailed signals.
Sixth, although the investment experiment in this study was
conducted with two stocks, there are many cases where three or
Frontiers in Artificial Intelligence | www.frontiersin.org 8May 2022 | Volume 5 | Article 865950
Taguchi et al. Constructing Equity Investment Strategies
more stocks are used in actual operations. In future research,
we would like to extend the investment model to multiple
dimensions and give priority to experiments with three or
more stocks. The following points are specific implementation
methods for portfolios of three or more stocks. First, we need
to ensure the diversification effect of stocks by using correlation
coefficients. Second, as the number of stocks increases, the timing
of rebalancing increases in proportion to the number of stocks, it
is necessary to obtain higher frequency data on text and stock
price data. It is also important to set a quantitative threshold
for the volatility of stocks. In addition, it is desirable to use
methods such as the mean-variance model proposed by Ma et al.
(2021) and Wang et al. (2020) when determining the maximum
investment ratio.
Finally, although the experiments in this study were limited
to stock investment strategies, we would like to extend
this study to future corporate bonds, Foreign Exchange and
virtual currencies.
7. CONCLUSION
In this study, we created a polarity index using BERT, performed
time series forecasting using BiGRU, and assigned a regime to
the forecasted values using GMM-HMM to create a signal for
portfolio rebalancing. Consequently, the strategy proposed in
this study proved effective, indicating that analyst reports are
useful for investment. In the future, we would like to create
a signal that replaces the architecture of the regime-switching
model with an anomaly detection algorithm. In addition, we
would like to conduct investment experiments with three or
more stocks in one portfolio since we have created two stocks
in one portfolio in this experiment. We seek to calculate the
investment ratio using reinforcement learning in conjunction
with this. Furthermore, we would like to conduct investment
simulations using polar indexes created by other financial texts
and compare the current results.
DATA AVAILABILITY STATEMENT
The data analyzed in this study is subject to the following
licenses/restrictions: this data can be used for academic
authorized researchers. Requests to access these datasets should
be directed to KH, kenji.hiramatsu@ifis.co.jp; HS, sakaji@sys.t.u-
tokyo.ac.jp.
AUTHOR CONTRIBUTIONS
RT, HW, HS, and KI contributed to the conception and design
of the study. KH organized the database. RT and HW performed
the statistical analysis. RT wrote the first draft of the manuscript
and sections of the manuscript. All authors contributed to
the revision of the manuscript and read and approved the
submitted versions.
FUNDING
This study was partially supported by JST-Mirai Program Grant
Number JPMJPI20B1, Japan. This study was not funded by IFIS
Japan Limited.
SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be found
online at: https://www.frontiersin.org/articles/10.3389/frai.2022.
865950/full#supplementary-material
REFERENCES
Asquith, P., Mikhail, M. B., and Au, A. S. (2005). Information content of
equity analyst reports. J. Fin. Econ. 75, 245–282. doi: 10.1016/j.jfineco.2004.
01.002
Chen, C.-C., Huang, H.-H., and Chen, H.-H. (2020). “Issues and perspectives
from 10,000 annotated financial social media data,” in Proceedings of the 12th
Language Resources and Evaluation Conference (Marseille: European Language
Resources Association), 6106–6110.
Chen, W., Zhang, H., Mehlawat, M. K., and Jia, L. (2021). Mean-variance portfolio
optimization using machine learning-based stock price prediction. Appl. Soft
Comput. 100, 106943. doi: 10.1016/j.asoc.2020.106943
Cho, K., van Merrienboer, B., Gülçehre, Ç., Bougares, F., Schwenk, H., and
Bengio, Y. (2014). Learning phrase representations using RNN encoder-
decoder for statistical machine translation. arXiv preprint arXiv:1406.1078.
doi: 10.3115/v1/D14-1179
Chung, J., Gulcehre, C., Cho, K., and Bengio, Y. (2014). Empirical evaluation
of gated recurrent neural networks on sequence modeling. arXiv preprint
arXiv:1412.3555. doi: 10.48550/ARXIV.1412.3555
Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019). BERT: pre-training
of deep bidirectional transformers for language understanding. arXiv preprint
arXiv:1810.04805. doi: 10.48550/ARXIV.1810.04805
Hiew, J. Z. G., Huang, X., Mou, H., Li, D., Wu, Q., and Xu, Y. (2019). BERT-based
financial sentiment index and LSTM-based stock return predictability. arXiv
preprint arXiv:1906.09024. doi: 10.48550/ARXIV.1906.09024
Hiramatsu, K., Kotaro, M., Sakai, H., and Sakaji, H. (2021). [Analyst report tone
information value] Anarisutorepoto no ton no Joho Kachi (in Japanese). Securit
Anal. J. 59, 86–97. Available online at: https://ci.nii.ac.jp/naid/40022479821/
Izumi, K., and Sakaji, H. (2019). “Economic causal-chain search using text mining
technology,” in Proceedings of the First Workshop on Financial Technology and
Natural Language Processing (Macao), 61–65.
Katayama, D., Kino, Y., and Tsuda, K. (2019). A method of sentiment polarity
identification in financial news using deep learning. Proc. Comput. Sci. 159,
1287–1294. doi: 10.1016/j.procs.2019.09.298
Liu, M., Huo, J., Wu, Y., and Wu, J. (2021). Stock market trend analysis
using hidden Markov model and long short term memory. arXiv preprint
arXiv:2104.09700.
Ma, Y., Han, R., and Wang, W. (2021). Portfolio optimization with return
prediction using deep learning and machine learning. Expert Syst. Appl. 165,
113973. doi: 10.1016/j.eswa.2020.113973
Sakaji, H., Kuramoto, R., Matsushima, H., Izumi, K., Shimada, T., and Sunakawa,
K. (2019). “Financial text data analytics framework for business confidence
indices and inter-industry relations,” in Proceedings of the First Workshop
on Financial Technology and Natural Language Processing (Macao, China),
40–46.
Sakaji, H., Sakai, H., and Masuyama, S. (2008). “Automatic extraction of basis
expressions that indicate economic trends,” in Advances in Knowledge Discovery
and Data Mining. PAKDD 2008, eds T. Washio, E. Suzuki, K. M. Ting, and A.
Inokuchi (Berlin; Heidelberg: Springer), 977–984. doi: 10.1007/978-3-540-681
25-0_102
Frontiers in Artificial Intelligence | www.frontiersin.org 9May 2022 | Volume 5 | Article 865950
Taguchi et al. Constructing Equity Investment Strategies
Suzuki, M., Sakaji, H., Izumi, K., Matsushima, H., and Ishikawa, Y. (2020).
Forecasting net income estimate and stock price using text mining from
economic reports. Information 11, 292. doi: 10.3390/info11060292
Taguchi, R., Watanabe, H., Hirano, M., Suzuki, M., Sakaji, H., Izumi, K., et al.
(2021). “Market trend analysis using polarity index generated from analyst
reports,” in 2021 IEEE International Conference on Big Data (Big Data)
(Orlando, FL), 3486–3494. doi: 10.1109/BigData52589.2021.9671702
Talha, S. A. W., Fleury, A., and Ambellouis, S. (2017). “Human action recognition
from body-part directional velocity using hidden Markov models,” in 2017 16th
IEEE International Conference on Machine Learning and Applications (ICMLA),
1035–1040. doi: 10.1109/ICMLA.2017.00-14
Wang, W., Li, W., Zhang, N., and Liu, K. (2020). Portfolio formation with
preselection using deep learning from long-term financial data. Expert Syst.
Appl. 143, 113042. doi: 10.1016/j.eswa.2019.113042
Yan, J., Liu, J., Yu, Y., and Xu, H. (2021). Water quality prediction in the Luan river
based on 1-DRCNN and BIGRU hybrid neural network model. Water 13, 1273.
doi: 10.3390/w13091273
Yono, K., Sakaji, H., Matsushima, H., Shimada, T., and Izumi, K. (2020).
Construction of macroeconomic uncertainty indices for financial market
analysis using a supervised topic model. J. Risk Fin. Manage. 13, 79.
doi: 10.3390/jrfm13040079
Yun, H., Lee, M., Kang, Y. S., and Seok, J. (2020). Portfolio management via
two-stage deep learning with a joint cost. Expert Syst. Appl. 143, 113041.
doi: 10.1016/j.eswa.2019.113041
Zhang, Z., Zohren, S., and Roberts, S. (2020). Deep learning for
portfolio optimization. J. Fin. Data Sci. 2, 8–20. doi: 10.3905/jfds.2020.
1.042
Conflict of Interest: KH is employed by IFIS Japan Limited.
The remaining authors declare that the research was conducted in the absence of
any commercial or financial relationships that could be construed as a potential
conflict of interest.
Publisher’s Note: All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their affiliated organizations, or those of
the publisher, the editors and the reviewers. Any product that may be evaluated in
this article, or claim that may be made by its manufacturer, is not guaranteed or
endorsed by the publisher.
Copyright © 2022 Taguchi, Watanabe, Sakaji, Izumi and Hiramatsu. This is an
open-access article distributed under the terms of the Creative Commons Attribution
License (CC BY). The use, distribution or reproduction in other forums is permitted,
provided the original author(s) and the copyright owner(s) are credited and that the
original publication in this journal is cited, in accordance with accepted academic
practice. No use, distribution or reproduction is permitted which does not comply
with these terms.
Frontiers in Artificial Intelligence | www.frontiersin.org 10 May 2022 | Volume 5 | Article 865950
