Analysis of Variance (ANOVA)

The sum of squares of a regression model is usually represented in the Analysis of Variance (ANOVA) table. The ANOVA table contains the sum of squares (SST, SSE, and SSR), the degrees of freedom, the mean squares (MSR and MSE), and F-statistics.

The typical format of ANOVA is as shown below:

$$\begin{array}{c|c|c|c|c} \textbf{Source} & \textbf{Sum of Squares} & \textbf{Degrees} & \textbf{Mean} & \textbf{F-statistic} \\ & & \textbf{of} & \textbf{square} & \\ & & \textbf{Freedom} & & \\ \hline { \text{Regression} \\ \text{(Explained)} } & SSR=\sum_{i=1}^{n}\left(\widehat{Y_i}-\bar{Y}\right)^2 & 1 & MSR=\frac{SSR}{1} & F=\frac{MSR}{MSE} \\ \hline { \text{Residual} \\ \text{(explained)} } & SSE=\sum_{i=1}^{n}\left(Y_i-\widehat{Y_i}\right)^2 & n-2 & MSE=\frac{SSE}{n-2} & \\ \hline \text{Total} & SST=\sum_{i=1}^{n}\left(Y_i-\bar{Y}\right)^2 & n-1 & & \end{array}$$

Standard Error of EstimateStandard Error of Estimate, $$S_e$$ or $$SEE$$, is referred to as the root mean square error or standard error of the regression. It measures the distance between the observed and dependent variables predicted by the regression model. The Standard Error of Estimate is easily calculated from the ANOVA table using the following formula:

$${\text{Standard Error of Estimate }}(S_e)=\sqrt{MSE}=\sqrt{\frac{\sum_{i = 1}^{n}\left(Y_i-\hat{Y}\right)^2}{n-2}}$$

The standard error of estimate, coefficient of determination, and F-statistic are the measures that can be used to gauge the goodness of fit of a regression model. In other words, these measures are used to tell the extent to which a regression model syncs with data.

The smaller the Standard Error of Estimate is, the better the fit of the regression line. However, the Standard Error of Estimate does not tell us how well the independent variable explains the variation in the dependent variable.

Example: Calculating and Interpreting F-Statistic

The completed ANOVA table for the regression model of the inflation rate against the unemployment rate over 10 years is given below:

$$\begin{array}{c|c|c|c|c} \textbf{Source} & \textbf{Sum of} & \textbf{Degrees of} & \textbf{Mean Sum} & \textbf{F-Statistic} \\ & \textbf{Squares} & \textbf{Freedom} & \textbf{of Squares} & \\ \hline \text{Regression} & 10.568 & 1 & 10.568 & ? \\ \hline \text{Error} & 3.136 & 8 & 0.392 \\ \hline \text{Total} & 13.704 & 9 \end{array}$$

1. Use the above ANOVA table to calculate the F-statistic.
2. Test the hypothesis that the slope coefficient is equal to a 5% significance level.

Solution

1. $$=\frac{{\text{Mean Regression Sum of Squares (MSR)}}}{{\text{Mean Squared Error(MSE)}}}=\frac{10.568}{0.392}=26.960$$
2. We are testing the null hypothesis $$H_0:b_1=0$$ against the alternative hypothesis $$H_1: b_1 \neq 0$$. The critical F-value for $$k = 1$$ and $$n-2 = 8$$ degrees of freedom at a 5% significance level is roughly 5.32. Note that this is a one-tail test, and therefore, we use the 5% F-table. Remember that the null hypothesis is rejected if the calculated value of the F-statistic is greater than the critical value of F. Since $$26.960 \gt 5.32$$, we reject the null hypothesis and conclude that the slope coefficient is significantly different from zero.Notice that we also rejected the null hypothesis in the previous examples. We did so because the 95% confidence interval did not include zero.An F-test duplicates the t-test in regard to the slope coefficient significance for a linear regression model with one independent variable. In this case, $$t^2={2.306}^2\approx 5.32$$. Since the F-statistic is the square of the t-statistic for the slope coefficient, its inferences are the same as the t-test. However, this is not the case for multiple regressions.

Question

Consider the following analysis of variance (ANOVA) table:

$$\begin{array}{c|c|c|c} \textbf {Source} & \textbf {Degrees of} & \textbf { Sum of } & \textbf {Mean Sum} \\ & \textbf{Freedom} & \textbf {Squares} & \textbf{of Squares} \\ \hline \text {Regression} & 1 & 1,701,563 & 1,701,563 \\ \hline \text {Error} & 3 & 106,800 & 13,350 \\ \text {(Unexplained)} & & & \\ \hline \text {Total} & 4 & 1,808,363 & \\ \end{array}$$

The value of $$R^2$$ and the F-statistic for the test of fit of the regression model are closest to:

1. 6% and 16.
2. 94% and 127.
3. 99% and 127.

Solution

The correct answer is B.

$$R^2=\frac{\text{Sum of Squares Regression (SSR)}}{\text{Sum of Squares Total (SST)}}=\frac{1,701,563}{1,808,363}=0.94=94\%$$

\begin{align*} F & =\frac{\text{Mean Regression Sum of Squares (MSR)}}{\text{Mean Squared Error (MSE)}} \\ & =\frac{1,701,563}{13,350} =127.46\approx 127 \end{align*}

Shop CFA® Exam Prep

Offered by AnalystPrep

Featured Shop FRM® Exam Prep Learn with Us

Subscribe to our newsletter and keep up with the latest and greatest tips for success
Shop Actuarial Exams Prep Shop Graduate Admission Exam Prep

Sergio Torrico
2021-07-23
Excelente para el FRM 2 Escribo esta revisión en español para los hispanohablantes, soy de Bolivia, y utilicé AnalystPrep para dudas y consultas sobre mi preparación para el FRM nivel 2 (lo tomé una sola vez y aprobé muy bien), siempre tuve un soporte claro, directo y rápido, el material sale rápido cuando hay cambios en el temario de GARP, y los ejercicios y exámenes son muy útiles para practicar.
diana
2021-07-17
So helpful. I have been using the videos to prepare for the CFA Level II exam. The videos signpost the reading contents, explain the concepts and provide additional context for specific concepts. The fun light-hearted analogies are also a welcome break to some very dry content. I usually watch the videos before going into more in-depth reading and they are a good way to avoid being overwhelmed by the sheer volume of content when you look at the readings.
Kriti Dhawan
2021-07-16
A great curriculum provider. James sir explains the concept so well that rather than memorising it, you tend to intuitively understand and absorb them. Thank you ! Grateful I saw this at the right time for my CFA prep.
nikhil kumar
2021-06-28
Very well explained and gives a great insight about topics in a very short time. Glad to have found Professor Forjan's lectures.
Marwan
2021-06-22
Great support throughout the course by the team, did not feel neglected
Benjamin anonymous
2021-05-10
I loved using AnalystPrep for FRM. QBank is huge, videos are great. Would recommend to a friend
Daniel Glyn
2021-03-24
I have finished my FRM1 thanks to AnalystPrep. And now using AnalystPrep for my FRM2 preparation. Professor Forjan is brilliant. He gives such good explanations and analogies. And more than anything makes learning fun. A big thank you to Analystprep and Professor Forjan. 5 stars all the way!
michael walshe
2021-03-18
Professor James' videos are excellent for understanding the underlying theories behind financial engineering / financial analysis. The AnalystPrep videos were better than any of the others that I searched through on YouTube for providing a clear explanation of some concepts, such as Portfolio theory, CAPM, and Arbitrage Pricing theory. Watching these cleared up many of the unclarities I had in my head. Highly recommended.