Paper: arXiv 2507.18643

Authors: Syeda Tasnim Fabiha, Rubaiyat Jahan Mumu, Farzana Aktar, B M Mainul Hossain

Abstract

Share market is one of the most important sectors of economic development of a country. Everyday almost all companies issue their shares and investors buy and sell shares of these companies. Generally investors want to buy shares of the companies whose market liquidity is comparatively greater. Market liquidity depends on the average price of a share. In this paper, a thorough linear regression analysis has been performed on the stock market data of Dhaka Stock Exchange. Later, the linear model has been compared with random forest based on different metrics showing better results for random forest model. However, the amount of individual significance of different factors on the variability of stock price has been identified and explained. This paper also shows that the time series data is not capable of generating a predictive linear model for analysis.

Complexity vs Empirical Score

  • Math Complexity: 2.0/10
  • Empirical Rigor: 4.0/10
  • Quadrant: Philosophers — conceptual discussion, lower math and empirical scores

Why this score: The paper uses standard regression and random forest without advanced mathematical derivations, but it applies these methods to real stock data with performance metrics, placing it in the low-math, low-rigor quadrant.

Research Flowchart

  flowchart TD
  A["Research Goal: Predict Share Market Prices for DSE"] --> B["Data Collection & Preparation"]
  B --> C["Methodology: Linear Regression & Random Forest"]
  C --> D{"Model Evaluation & Comparison"}
  D --> E["Linear Regression Model"]
  D --> F["Random Forest Model"]
  E --> G["Outcomes: Low R2 for Linear Model<br>Time Series Unsuitable for Linear Prediction"]
  F --> H["Outcomes: Better Performance for Random Forest<br>Identified Key Market Factors"]
  G & H --> I["Conclusion: Factors Identified, RF Superior for Prediction"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a predictive analysis model for Dhaka Stock Exchange prices using linear regression and random forest methods.

Source sentence
In this study we propose to run a predictive analysis for stock market price of Dhaka Stock Exchange.

Assumes 5

  • Linear regression model assumes measurement‑error‑free observations, linearity, normality, homoscedasticity, no multicollinearity and no autocorrelation. model
    Source sentence
    model satisfies the conditions of linear modelling such as measurement error free observation of predictors and responses, linearity, normality, homoscedasticity, no multicollinearity and no autocorrelation.
  • Predictors are treated as important contributors initially before selection of best predictors. data
    Source sentence
    Initially all the predictors have been considered as important contributors for the model generation. Later, analysis is performed on the identification of the best predictors, which contribute most for describing the variability of the stock closing prices.
  • The dataset is a quarterly time‑series from 2013 to 2017 for five pharmaceutical companies. data
    Source sentence
    Data was collected on quarter basis from 4th quarter of 2013 to 1st quarter of 2017. The five selected companies are: ACI Pharmaceuticals, Beximco Pharmaceuticals Ltd, Renata Limited, Glaxo SmithKline Bangladesh Ltd and Square Pharmaceuticals Limited.
  • Random forest improves accuracy by bagging multiple trees, sacrificing interpretability. model
    Source sentence
    In order to avoid the high variance problem of decision trees, bagging increases the prediction accuracy at the expense of loss of interpretability.
Show 1 more
  • Time‑series data has inherent collinearity among predictors, favoring tree‑based methods. statistical
    Source sentence
    Because of this, the model is more biased towards having better performance using tree based method random forest.

Data

Backtest sample: Dhaka Stock Exchange · equities · daily

  • Dhaka Stock Exchange quarterly data (2013‑2017) · Lanka Bangla Financial Portal, Dhaka Stock Exchange Banglade fundamentals public
    2013-2017 (quarterly) · Five pharmaceutical companies
    Source sentence
    Most of the data was collected from the Stock market related websites of Bangladesh like Lanka Bangla Financial Portal, Dhaka Stock Exchange Bangladesh Ltd. and Trading Economics.

Costs

Not applicable The paper reports no trading or execution result.

Finds 2

  • Linear regression identifies Term, Total Asset Turnover, Debt‑to‑Total‑Asset and company sector as significant predictors of daily stock price. statistical
    Source sentence
    At 0.001 confidence level, Term, TATO, DTA are Panel are related to the response. This means that, time sequence of stock data, total asset turnover of a company, Debt on total asset and company type impact the stock price of a day.
  • ROE has a positive coefficient, while Current Ratio has a negative coefficient in the linear model. statistical
    Source sentence
    ... The estimate of CR is negative also explains the fact clearly that having large values for CR decreases stock price of shares.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 17 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (38)

  1. Zur Elektrodynamik bewegter K "o rper . ( German ) [ On the electrodynamics of moving bodies] (1905)
  2. ANNUAL FLOOD REPORT 2014 (2014)
  3. Multi-variate flood damage assessment: a tree-based data-mining approach (2013)
  4. Flood damage and influencing factors: New insights from the August 2002 flood in Germany (2005)
  5. Modeling flood damage: case of Tokai Flood 2000 (2005)
  6. Flood hazard and risk analysis in the southwest region of Bangladesh (2005)
  7. Evaluating flood hazard for land-use planning in Greater Dhaka of Bangladesh using remote sensing and GIS techniques (2007)
  8. Estimation of flood damage functions for river basin planning: a case study in Bangladesh (2015)
and 30 more
  1. Factors of influence on flood damage mitigation behaviour by households (2014)
  2. Long-term development and effectiveness of private flood mitigation measures: an analysis for the German part of the river Rhine (2012)
  3. Perception and communication of flood risks: a systematic review of empirical research (2013)
  4. Floods and poverty traps: Evidence from Bangladesh (2007)
  5. Coping with floods: preparedness, response and recovery of flood-affected residents in Germany in 2002 (2007)
  6. Distribution of impacts of natural disasters across income groups: A case study of New Orleans (2007)
  7. What determines vulnerability to floods; a case study in Georgetown, Guyana (1997)
  8. Flood damage, vulnerability and risk perception–challenges for flood damage research (2006)
  9. Methods and criteria for model selection (2004)
  10. The assumptions of the linear regression model (1971)
  11. Exploratory data analysis (1977)
  12. Testing for serial correlation in least squares regression. II (1951)
  13. Random forests (2001)
  14. Boosted trees for ecological modeling and prediction (2007)
  15. The perceptron: A probabilistic model for information storage and organization in the brain. (1958)
  16. Learning internal representations by error propagation (1985)
  17. Regression diagnostics; identifying influential data and sources of collinearity (1980)
  18. Regression analysis by example (1977)
  19. Factors Affecting the Stock Price Variability of Dhaka Stock Exchange (DSE) (2014)
  20. Stock market prediction using machine learning techniques (2016)
  21. Stock market prediction system with modular neural networks (1990)
  22. Factors Affecting the Stock Price Movement: A Case Study on Dhaka Stock Exchange (2015)
  23. (12 October 2011) Bangladesh starts market rescue fund — link
  24. Exploratory data analysis (1977)
  25. LankaBangla Financial Portal — link
  26. Dhaka Stock Exchange Ltd. — link
  27. Trading Economics — link
  28. Time series analysis (1994)
  29. Regression and time series model selection in small samples (1989)
  30. Problems That May Occur in Time Series Multiple Regression — link

0 of 38 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.