how to make a residual plot ti 84

Mastering Ti 84: Step-by-Step Guide to Creating Residual Plots

By Brian Young
AD

Are you a TI-84 user eager to delve into the world of data visualization and regression analysis? One of the most powerful tools at your disposal is the Residual Plot. This graphical method allows you to compare the differences between your observed and predicted values, providing valuable insights into the accuracy of your regression model. So, let's explore how to create a Residual Plot on your TI-84 calculator.

Before we dive in, ensure your TI-84 is up to date with the latest operating system. This guide assumes you're using a standard TI-84 Plus or TI-84 Plus Silver Edition calculator, with version 2.53MP or later. Now, let's get started!

Preparing Your Data

For our example, we'll use a simple set of (x, y) data points: (1, 5), (2, 7), (3, 9), (4, 11), and (5, 13). First, store your data in L1 and L2 lists:

AD
PLOTTING IN R - Your First Steps
PLOTTING IN R - Your First Steps
Log Transformations with TI-84
Log Transformations with TI-84
an info sheet with different colors and numbers on it, including the text that says plot points
an info sheet with different colors and numbers on it, including the text that says plot points

L1={1,2,3,4,5}
L2={5,7,9,11,13}

Next, ensure your calculator is in 'Stats' mode by pressing MODE and selecting 'Statistics'.

Unlocking Data from Graphs: How to Digitise Plots and Figures with WebPlotDigitizer | Towards Data Science
Unlocking Data from Graphs: How to Digitise Plots and Figures with WebPlotDigitizer | Towards Data Science
Box Plot
Box Plot

Calculate the Regression Line and comorbidities

The first step in creating a Residual Plot is to calculate the regression line equation (slope 'a' and intercept 'b').

Calculate the Regression Line

Press STAT then 'CALC', and choose 'LinReg (ax+b)' under the 'yen' icon. Enter L1 and L2, and press ENTER to display the regression line equation (ax+b) and the correlation coefficient, 'r'.

AD

If you're not familiar with the regression line equation, don't worry! The calculator will display it in the format 'y=ax+b', making it easy to understand.

Create a List of Residuals

Now that we have the regression equation, we can calculate the residuals (the difference between the observed and predicted y-values).

Create a new list, L3, to store the residuals:

AD

L3=L2-L1*a-b

Here's what we're doing: subtract the predicted y-values (L1*a+b) from the observed y-values (L2), then store the results in L3.

The residuals represent the vertical distance between the observed data points and the regression line. The sign of the residual indicates whether the observed value is above or below the regression line.

AD

Creating the Residual Plot

Now that we have our residuals, it's time to create the plot!

Plot the Data Points

Press STAT then 'PLOT', and choose 'Scatter'. Enter L1 and L2, and press ENTER. You'll see a scatter plot of your data points.

But wait, where's the regression line? We'll add that next.

Plot the Regression Line

Press PLOT again, and choose 'Line'. Enter the transformed equations i(L1)*a+b and 0, and press ENTER. This will create the regression line on your scatter plot, based on the equation we calculated earlier.

IMPORTANT: Be sure to use 'i(L1)' to represent the transformed x-values. This ensures the proper slope and intercept are used for the line.

Add the Residuals to the Plot

Finally, let's add the residuals to our plot. Press PLOT one more time, and choose 'Resize'. This allows you to adjust the scale of your y-axis to accommodate the residuals.

Enter the following settings:

  • Xmin = -5
    Xmax = 15
    Ymin = -10
    Ymax = 10
    Xres = .05
    Yres = .1

Press ENTER, then re-enter your data for L1, L2, and L3. Press GRAPH to create your Residual Plot.

Congratulations! You've just created your first Residual Plot on your TI-84 calculator. This powerful tool can help you identify and investigate potential issues with your regression model, such as heteroscedasticity or outliers.

Now that you're a Residual Plot master, why not try creating your own plots using different datasets? Keep experimenting, and happy calculating!

AD