Quantitative Methods (M34EFA): BAE Systems Stock Return Analysis

This report analyzes the stock returns of BAE Systems using quantitative methods, specifically linear regression. Monthly price data from February 2001 to April 2017 was used to compute market returns and BAE system returns, with 3-month treasury bill rates also factored in. The analysis reveals a significant linear association between the company's return and its relative risk compared to the market. The fitted linear regression model, Rt = 366.8531 – 0.5076(Rmt-Rft), is significant, with an r-squared value of 0.470, indicating that the model accounts for 47% of the variation in the company's return. The Durbin-Watson value suggests no correlation between residuals, making the model suitable for predicting BAE Systems' returns. The independent variable coefficient shows that a one-unit increase in relative risk leads to a 0.5076 reduction in the stock's rate of return.

Quantitative Methods for Accountants
Name
The Name of the Class (Course)
Professor (Tutor)
The Name of the School (University)
The City and State where it is located
Date

Question 4
The BAE system data were obtained from the website: https://uk.investing.com/equities/bae-
systems-historical-data, for which they were available as from 01-04-2004 to-date. On the
monthly 3-month treasury bill rates, I used the end of the End month level of the discount
rate, 3-month Treasury bills, Sterling. The following variables were computed as per the
instructions. That is; Market returns → Rmt = ln(ABS(Pmt/ Pmt-1))*100, an absolute was used
since the logarithmic could not be found for the negative values. BAE system returns → Rt =
ln(ABS(Pt/ Pt-1))*100. The 3-month treasury bills were also converted from the annual rate of
return to monthly using the function RMrf = RArf/12.
A linear regression model was fitted to the data, to obtain a linear model of the form,
Rt =α + β ( Rmt −Rft )
The results of the analysis are as follows.
Table 1: Model summary
Regression Analysis
r² 0.470 n 157
r -0.686 k 1
Std. Error 71.147 Dep. Var. Rt
Table 2: ANOVA summary of the regression model
ANOVA
table
Source SS df MS F p-value
Regression 696,735.9831 1
696,735.983
1 137.64 3.77E-23
Residual 784,604.1302 155 5,061.9621
Total
1,481,340.113
3 156
Table 3: Regression coefficient summary
Regression output confidence interval
Variables coefficients std. error t (df=155) p-value
95%
lower
95%
upper
Intercept 366.8531 11.0691 33.142 3.04E-72 344.9873 388.7189
Rmt-Rft -0.5076 0.0433 -11.732 3.77E-23 -0.5931 -0.4222
The fitted linear regression model is Rt = 366.8531 – 0.5076(Rmt-Rft).

This fitted model is significant (F (1, 155) = 137.64, p-value < .001) (Chatterjee & Hadi.,
2015). This means that there is a linear association between the Rt and the independent
variable (Rmt-Rft). The summary shows that as the company’s risk and return relative to the
market risk increases the return of the company decreases. The r-squared value 0.470,
indicating that the model could take into account 47.0% of the sources of variation of the
company’s return when the relative risk of the company relative to the market risk is known.
The Durbin-Watson value is 1.83, which is close to 2, suggesting that there is absolutely no
correlation between the residuals (Heiberger & Holland, 2015). Therefore, the fitted model is
ideal for predicting the return of the BEA systems.
The independent variable coefficient shows that when the relative risk of the company
relative to the market risk (Rmt-Rft) increases by one unit, the rate of return of the stock will
reduce by 0.5076.

Bibliography
Chatterjee, S. & Hadi., A. S., 2015. Regression analysis by example. s.l.:John Wiley & Sons.
Heiberger, R. M. & Holland, B., 2015. Multiple Regression—Regression Diagnostics.
Statistical Analysis and Data Display, pp. 345-375.