After fitting a regression curve, use it to predict a value or to solve for when something happens.
In short: Using a fitted model to make predictions means substituting a value into the regression equation to estimate an output. Interpolation, predicting inside the data range, is usually reliable, while extrapolation beyond it is less so. Store the regression equation in a function slot so you can evaluate it quickly and graph it with the data.
| TI-84 Plus CE | Casio fx-CG50 and fx-CG100 | TI-Nspire CX | |
|---|---|---|---|
| Key sequence | Paste the equation into Y1 (VARS → Statistics → EQ → RegEQ), then use a table or solve Y1 = value. | Copy the regression equation to the graph (Y=), then use G-Solv or a table to read predictions. | Store the regression equation to f1(x), then evaluate f1(a) or solve f1(x) = value. |
Tip: Solve model = target to find when a quantity reaches a given level (e.g. when sales hit 1000).
Tip: Solve model = target to find when a quantity reaches a given level (e.g. when sales hit 1000).
Tip: Solve model = target to find when a quantity reaches a given level (e.g. when sales hit 1000).
Here's a real IB-style question that uses exactly this technique.
A model gives y = 3(2)ˣ. Use it to predict y when x = 6.
After fitting a regression curve, use it to predict a value or to solve for when something happens. You've already fitted a regression model and now need a value it predicts. Substituting an x-value into the stored regression equation, rather than retyping it.
1. Fit the model so its equation is available to reuse. 2. Paste the equation into Y1 (VARS → Statistics → EQ → RegEQ), then use a table or solve Y1 = value. 3. Predict within the data range (interpolation); extrapolating beyond it is unreliable.
1. Fit the model so its equation is available to reuse. 2. Copy the regression equation to the graph (Y=), then use G-Solv or a table to read predictions. 3. Predict within the data range (interpolation); extrapolating beyond it is unreliable.
1. Fit the model so its equation is available to reuse. 2. Store the regression equation to f1(x), then evaluate f1(a) or solve f1(x) = value. 3. Predict within the data range (interpolation); extrapolating beyond it is unreliable.
Solve model = target to find when a quantity reaches a given level (e.g. when sales hit 1000).
In the worked example on this page (2 marks, Paper 2), the marks are: M1: Substitute x = 6 into the stored regression equation; A1: 192.
Questions that need this technique link back to this guide. Try one in practice mode, or see every GDC guide.