X is the independent variables data matrix, such that each column represents one variable.

Mask is the boolean array to choose the explanatory variables in the model. If missing, all variables in X are included.

Y is the response or the dependent variable data array (one dimensional array of cells (e.g. rows or columns)).

Intercept is the constant or the intercept value to fix (e.g. zero). If missing, an intercept will not be fixed and is computed normally.

Return_type is a switch to select the return output (1 = SSR (default), 2 = SSE, 3 = SST, ... (see help file)).

1. The underlying model is described here.
2. $$\mathbf{y} = \alpha + \beta_1 \times \mathbf{PC}_1 + \dots + \beta_p \times \mathbf{PC}_p$$
3. The regression ANOVA table examines the following hypothesis:
   
   $$\mathbf{H}_o: \beta_1 = \beta_2 = \dots = \beta_p = 0 $$
   $$\mathbf{H}_1: \exists \beta_i \neq 0, i \in \left[1,0 \right ] $$
4. In other words, the regression ANOVA examines the probability that the regression does NOT explain the variation in $\mathbf{y}$, i.e. that any fit is due purely to chance.
5. The MLR_ANOVA calculates the different values in the ANOVA tables as follows:
   
   $$\mathbf{SST}=\sum_{i=1}^N \left(Y_i - \bar Y \right )^2 $$
   $$\mathbf{SSR}=\sum_{i=1}^N \left(\hat Y_i - \bar Y \right )^2 $$
   $$\mathbf{SSR}=\sum_{i=1}^N \left(Y_i - \hat Y_i \right )^2 $$
   Where:
   • $\mathbf{PC}$ is the principal component.
   • $N$ is the number of non-missing observations in the sample data.
   • $\bar Y$ is the empirical sample average for the dependent variable.
   • $\hat Y_i$ is the regression model estimate value for the i-th observation.
   • $\mathbf{SST}$ is the total sum of squares for the dependent variable.
   • $\mathbf{SSR}$ is the total sum of squares for the regression (i.e. $\hat y$) estimate.
   • $\mathbf{SSE}$ is the total sum of error (aka residuals $\epsilon$) terms for the regression (i.e. $\epsilon = y - \hat y$) estimate.
   • $\mathbf{SST} = \mathbf{SSR} + \mathbf{SSE}$
   AND
   $$\mathbf{MSR} = \frac{\mathbf{SSR} }{p} $$
   
   $$\mathbf{MSE} = \frac{ \mathbf{SSE} }{N-p-1}$$
   
   $$\mathbf{F-Stat} = \frac{\mathbf{MSR} }{\mathbf{MSE} }$$
   
   
   Where:
   • $p$ is the number of explanatory (aka predictor) variables in the regression.
   • $\mathbf{MSR}$ is the mean squares of the regression.
   • $\mathbf{MSE}$ is the mean squares of the residuals.
   • $\textrm{F-Stat}$ is the test score of the hypothesis.
   • 
     $\textrm{F-Stat} \sim \mathbf{F}\left(p,N-p-1 \right)$
6. The sample data may include missing values.
7. Each column in the input matrix corresponds to a separate variable.
8. Each row in the input matrix corresponds to an observation.
9. Observations (i.e. rows) with missing values in X or Y are removed.
10. The number of rows of the response variable (Y) must be equal to the number of rows of the explanatory variable (X).
11. The MLR_ANOVA function is available starting with version 1.60 APACHE.

Files Examples

Related Links

• Wikipedia - Linear regression
• Wikipedia - Regression analysis
• Wikipedia - Ordinary least squares

References

• Hamilton, J .D.; Time Series Analysis , Princeton University Press (1994), ISBN 0-691-04289-6
• Kenney, J. F. and Keeping, E. S. (1962) "Linear Regression and Correlation." Ch. 15 in Mathematics of Statistics, Pt. 1, 3rd ed. Princeton, NJ: Van Nostrand, pp. 252-285