91Ó°ÊÓ

The Least Squares Method is a popular statistical technique used in regression analysis. It aims to find the best-fitting line through a set of data points by minimizing the sum of the squares of the vertical deviations from each point to the line.
In simple terms, the deviations are the differences between the observed values and those predicted by the line. By squaring these differences, we focus on the magnitude rather than the direction, allowing us an unbiased estimation.
The process to find the line, called the least squares regression line, involves calculating two key components:

• The **slope (b)**, which represents how much the dependent variable is expected to increase when the independent variable increases by one unit.
• The **y-intercept (a)**, which indicates the expected value of the dependent variable when the independent variable is zero.

In our exercise, we found that the least squares line was given by the equation \( y = -28.08 + 1.0483x \). This indicates that for each increase of 1 MPa in cube strength \( x \), we expect the axial strength \( y \) to increase by about 1.0483 MPa.

The Coefficient of Determination, commonly denoted as \( R^2 \), tells us how well our regression line predicts actual data points. It's a value ranging from 0 to 1, where:

• 0 indicates that the model does not explain any of the variability in the response variable.
• 1 suggests that the model explains all the variability.

Mathematically, \( R^2 \) is calculated by dividing the Regression Sum of Squares (\( SS_{reg} \)) by the Total Sum of Squares (\( SS_{tot} \)). In this exercise, we calculated \( R^2 \) as 0.0805, showing that around 8.05% of the variability in axial strength is explained by the cube strength.
A low \( R^2 \), like in our result, suggests that the model may not capture all influencing factors, indicating room for improvement. However, it might still be helpful if it at least partially explains the variations of the dependent variable.

Estimating the Standard Deviation of the residuals, or the Standard Error, evaluates the accuracy of predictions made by the regression line. It measures the average distance that the observed values fall from the regression line.
The formula used is:\[ S_{e} = \sqrt{\frac{SS_{res}}{n-2}} \]where \( SS_{res} \) is the Residual Sum of Squares and \( n \) is the number of data points.
In our case, the calculated standard error was 11.14 MPa, illustrating the typical deviation of the observed values from the predicted values in the context of our dataset.
A smaller standard error indicates a closer fit between the observed data points and the regression line, implying a more accurate model.

Interpreting Regression Coefficients is essential to understand the relationship between variables in a regression model. Each coefficient quantifies the change in the dependent variable due to a one-unit change in an independent variable.
For our linear regression, the equation \( y = -28.08 + 1.0483x \) has two coefficients:

• The **slope (1.0483)** shows how much \( y \), the axial strength, increases when \( x \), the cube strength, increases by one unit.
• The **intercept (-28.08)** is the expected axial strength when the cube strength is zero. Although it may not be practically useful in this context, as negative strength isn't possible, it is necessary for the line equation.

By analyzing these coefficients, we can make predictions and understand relationships between the variables in our dataset. They help reveal the underlying trend and facilitate decision-making based on the strength of the relationship depicted by the slope.