Paper: arXiv 2403.13192
Authors: Dennis Lartey Quayesam, Anani Lotsi, Felix Okoe Mettle
Abstract
Modeling financial data often relies on assumptions that may prove insufficient or unrealistic in practice. The Geometric Brownian Motion (GBM) model is frequently employed to represent stock price processes. This study investigates whether the behavior of weekly and monthly returns of selected equities listed on the Ghana Stock Exchange conforms to the GBM model. Parameters of the GBM model were estimated for five equities, and forecasts were generated for three months. Evaluation of estimation accuracy was conducted using mean square error (MSE). Results indicate that the expected prices from the modeled equities closely align with actual stock prices observed on the Exchange. Furthermore, while some deviations were observed, the actual prices consistently fell within the estimated confidence intervals.
Complexity vs Empirical Score
- Math Complexity: 6.5/10
- Empirical Rigor: 4.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper employs advanced mathematical concepts like stochastic differential equations and Itô’s lemma, but its empirical rigor is limited by small sample size (15 equities), lack of robust backtesting, and reliance on simple metrics like MSE without addressing transaction costs or out-of-sample performance beyond a three-month forecast.
Research Flowchart
flowchart TD A["Research Goal:<br>Do GSE returns conform to GBM?"] --> B["Data Selection<br>Weekly & Monthly Returns<br>5 Selected Equities"] B --> C["Parameter Estimation<br>Geometric Brownian Motion<br>Forecast Generation 3 Months"] C --> D["Model Evaluation<br>Mean Square Error (MSE)<br>Actual vs Expected Prices"] D --> E["Key Findings<br>Prices align with GBM model<br>Actual prices within confidence intervals"]
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 investigates whether weekly and monthly returns of selected Ghana Stock Exchange equities follow a Geometric Brownian Motion model and generates three‑month forecasts.
Source sentence
This study investigates whether the behavior of weekly and monthly returns of selected equities listed on the Ghana Stock Exchange conforms to the GBM model.
Assumes 3
- Weekly and monthly returns are stationary. statistical
Source sentence
The p-value obtained from the Augmented Dickey-Fuller Test for both the weekly and monthly returns of the GCB equity is below the 0.05 significance level. Consequently, we reject the null hypothesis of a unit root (non-stationarity) at the 5% significance level, leading to the conclusion that both the weekly and monthl
- Returns are normally distributed (for monthly returns). statistical
Source sentence
To validate whether the returns conform to a normal distribution, we proceed with the Shapiro-Wilk Test.
- Log‑returns follow a Geometric Brownian Motion (GBM) process. model
Source sentence
The stock price process is represented by the stochastic differential equation dp = µpdt + σpds
Data
- 15 selected equities listed on the Ghana Stock Exchange market-prices unknown
Source sentence
The methods proposed above are applied to 15 selected equities listed on the Ghana Stock Exchange.
- Ghana Commercial Bank (GCB) equity weekly and monthly returns market-prices unknown
Source sentence
We specifically choose Ghana Commercial Bank (GCB) equity for demonstration and display its weekly and monthly returns results.
Costs
Not applicable
Finds 3
- Expected prices from the GBM model closely align with actual stock prices and all observed prices fall within the 95% confidence interval. predictive-accuracy
Source sentence
Results indicate that the expected prices from the modeled equities closely align with actual stock prices observed on the Exchange.
- All observed actual prices for various months fall within the 95% confidence interval. statistical
Source sentence
Nonetheless, all observed actual prices for various months fall within the 95%confidence interval.
- GBM assumptions were violated for 13 weekly returns but satisfied for monthly returns of five equities. negative
Source sentence
The study revealed that among the selected equities, at least one assumption of the Geometric Brownian Motion (GBM) was violated for 13 weekly returns, while the monthly returns of five equities met all GBM assumptions.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 18 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 (13)
- A methodology for stochastic analysis of share prices as Markov chains with finite states (2014)
- Comparing South African financial markets behaviour to the Geometric Brownian motion process (2008)
- A review on Geometric Brownian Motion in forecasting the share prices in Bursa Malaysia (2012)
- Continuous Time Processes in Times of Crisis: The Case Of GBM and CEV Models (2011)
- Development of modified geometric Brownian motion models by using stock price data and basic statistics (2009)
- Analysis of financial time series (2005)
- Option pricing when underlying stock returns are discontinuous (1976)
- Stock price modelling: Theory and Practice (2006)
and 5 more
- Determining the return volatility of the Ghana stock exchange before and during the COVID-19 pandemic using the exponential GARCH model (2022)
- Poisson Process Modeling of Pure Jump Equities on the Ghana Stock Exchange (2022)
- Exchange Rate and Stock Price Nexus: Evidence from Ghana (2021)
- Bayesian Switching Volatility Models for Analysing Stock Returns in Ghana (2022)
- Drivers of stock prices in Ghana: an empirical mode decomposition approach (2021)
0 of 13 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.