Summary (Grade 12 NSC Matric Mathematics): Revision Notes

Summary

Curve fitting and data analysis

Curve fitting is the process of fitting mathematical functions to sets of data points. This fundamental technique allows us to find patterns in data and make predictions based on observed relationships.

Intuitive curve fitting involves visually interpreting data by examining whether points on a scatter plot appear to follow a particular pattern. You might recognise linear, exponential, quadratic, or other function shapes when looking at the data distribution.

Line of best fit

The line of best fit (also called a trend line) is a straight line drawn through data points that best represents the overall pattern of the data. This line:

• Minimises the distance between itself and all data points
• Allows for estimation of missing data values
• Provides a mathematical model for the relationship between variables

Interpolation and extrapolation

Interpolation is the technique used to predict values that fall within the range of your available data. When you use interpolation, you're making predictions between known data points, which tends to be more reliable.

Extrapolation is the technique used to predict values beyond the range of your available data. This involves extending your trend line past the known data points. Extrapolation is generally less reliable than interpolation because you're making assumptions about patterns continuing outside the observed range.

Linear regression analysis

Linear regression analysis is a statistical technique used to determine exactly which linear function provides the best fit for a given set of data. Rather than drawing a line by eye, this method uses mathematical calculations to find the optimal line.

Least squares method

The least squares method provides an algebraic approach to finding the linear regression equation. This method minimises the sum of squared differences between observed and predicted values.

The linear regression equation has the form:

$\hat{y} = a + bx$

Where:

• $\hat{y}$ (y-hat) is the predicted value of y
• $a$ is the y-intercept
• $b$ is the slope of the line
• $x$ is the independent variable

Calculating the slope (b):

$b = \frac{n\sum x_i y_i - \sum x_i \sum y_i}{n\sum (x_i)^2 - (\sum x_i)^2}$

Calculating the y-intercept (a):

$a = \frac{1}{n}\sum y_i - \frac{b}{n}\sum x_i = \bar{y} - b\bar{x}$

This can also be written as: $a = \text{mean of y-values} - b \times \text{mean of x-values}$

Linear correlation coefficient

The linear correlation coefficient (denoted as r) measures both the strength and direction of the linear relationship between two variables.

The formula for the correlation coefficient is:

$r = b\left(\frac{\sigma_x}{\sigma_y}\right)$

Where:

• $b$ is the slope from the regression equation
• $\sigma_x$ is the standard deviation of x-values
• $\sigma_y$ is the standard deviation of y-values

Understanding correlation values:

The correlation coefficient always falls within the range $r \in [-1, 1]$:

Interpreting correlation strength:

• Values close to -1 or 1 indicate strong correlation
• Values close to 0 indicate weak correlation
• The sign (positive or negative) indicates the direction of the relationship