Fit a model (quadratic, cubic, exponential, power, sinusoidal)
Pick a curve that suits your data, let the calculator find its equation, compare candidates with R² and predict from the result. Every route is given for the TI-84 Plus CE, Casio fx-CG50, Casio fx-CG100 and TI-Nspire CX, followed by how to set the answer out.
A regression command turns a table of data into an equation. Put the inputs in one list and the outputs in a second, then run the command for the curve you think fits: quadratic, cubic, exponential, power or sinusoidal. You get the parameters of the equation, plus R² for every model except the sinusoidal one. The scatter diagram and the context should suggest the curve; R² helps you choose between close candidates.
Example
The number of users of an app, thousand, is recorded years after its launch.
: 0, 1, 2, 3, 4, 5, 6, 7
: 12, 18, 25, 36, 49, 70, 98, 141
(a) Fit a model of the form and state . (b) Estimate the number of users after 4.5 years.
Exponential regression on the two lists gives (a) with , a very close fit, and (b) roughly thousand users at 4.5 years, an estimate made between the first and last years recorded.
Choose your calculator
How to fit a model on a TI-84 Plus CE
Turn on R² (once) | |
R² stays hidden until diagnostics are switched on. Open the mode screen with mode, go down to the STAT DIAGNOSTICS line, choose ON with ▶ and confirm with enter. Leave with 2nd mode (QUIT). The setting lasts until the calculator is reset. |
|
Enter the data | |
Open the list editor with stat and 1:Edit. To empty a list, highlight its name (L1 or L2) at the top and press clear, then enter. |
|
Key the inputs into L1 one per line, with enter after each: 0, 1, 2, 3, 4, 5, 6, 7. Move across to L2 with ▶ and key the outputs the same way: 12, 18, 25, 36, 49, 70, 98, 141. The two lists need equal lengths. |
L1 = 0, 1, 2, 3, 4, 5, 6, 7 L2 = 12, 18, 25, 36, 49, 70, 98, 141 |
Choose the model and run it | |
Press stat and move to CALC with ▶. The curve fits are numbered in the menu: 5:QuadReg quadratic, |
|
For the example choose 0:ExpReg. With the Stat Wizards on, a screen asks for Xlist and Ylist: set Xlist to L1 and Ylist to L2 (press 2nd 1 for L1 and 2nd 2 for L2), then move to Calculate and press enter. If the Stat Wizards are off, type |
ExpReg y=a*b^x a=12.4068849 b=1.4142966 r²=0.9994408 r=0.9997204 |
Read a and b, and r² (the R² value). The same steps work for the other models; only the menu item changes. and , with the coefficients to 3 significant figures. |
|
Compare two models with R² | |
Run another model on the same lists: press stat, ▶ to CALC and choose 5:QuadReg, with Xlist L1 and Ylist L2 as before. The screen shows a, b, c and R². Compare the R² values of the models, then look at the graph before you decide. and . |
QuadReg y=ax²+bx+c a=2.8392857 b=-2.6011905 c=15.5416667 R²=0.9936753 |
Predict a value from the model | |
Before you run the regression, move to Store RegEQ on the wizard screen, press vars, then ▶ to Y-VARS and choose 1:Function then 1:Y1. The full-accuracy equation is then saved in Y1. On the home screen type when for the exponential model, to 1 decimal place. |
Y1(4.5) 59.0330102 |
A sinusoidal model | |
Enter the times in L1 and the depths in L2 (clear the old values first): L1 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24; L2 3.1, 5.6, 7, 6.4, 3.9, 1.6, 1.3, 3.4, 5.9, 7.1, 6.2, 3.8, 1.9. Then press stat, ▶ to CALC and choose C:SinReg. Leave Period empty when the times are equally spaced and in order, otherwise type a period guess. Move to Calculate and press enter. |
SinReg y=a*sin(bx+c)+d a=2.9173012 b=0.4526533 c=-0.3835108 d=4.2119403 |
The screen gives a, b, c and d and no R². The angle is always in radians, so put the calculator in radian mode before you use the stored equation. , with the coefficients to 3 significant figures. |
|
If R² does not appear, diagnostics are off: turn STAT DIAGNOSTICS on in the mode screen and run the regression again. Use the stored equation in Y1 rather than rounded coefficients when you predict.
How to fit a model on a Casio fx-CG50
Enter the data | |
From MENU open Statistics, which starts in the List Editor. |
|
Fill List 1 with the inputs, EXE after each value: 0, 1, 2, 3, 4, 5, 6, 7. Then fill List 2 with the outputs the same way: 12, 18, 25, 36, 49, 70, 98, 141. The lists need equal lengths. |
List1 = 0, 1, 2, 3, 4, 5, 6, 7 List2 = 12, 18, 25, 36, 49, 70, 98, 141 |
Open F2 (CALC) and F6 (SET). 2Var XList should read List 1 and 2Var YList List 2. Leave the screen with EXIT. |
|
Choose the model and run it | |
Press F2 (CALC) then F3 (REG). The first menu holds F3 (X²) for quadratic and F4 (X³) for cubic. Press F6 for the next menu, which holds Exp (F2), Power (F3) and Sin (F4). |
|
For the example press F6, then F2 (Exp), then F2 (abx). The results screen shows a, b, r, r² and MSe. Choose abx for the form , not aebx. and , with the coefficients to 3 significant figures. |
Exponential Reg(abx) a=12.4068849 b=1.4142966 r=0.9997204 r²=0.9994408 MSe=... |
Compare two models with R² | |
Press EXIT to go back to the List Editor and run another model on the same lists, for example F2 (CALC), F3 (REG), F3 (X²). Compare the r² values of the models, then look at the graph before you decide. and . |
Quadratic Reg a=2.8392857 b=-2.6011905 c=15.5416667 r²=0.9936753 MSe=... |
Predict a value from the model | |
With the results of the model on the screen, press F6 (DRAW). Then press SHIFT F5 (G-SOLVE) and F1 (Y-CAL), press EXE to choose the graph, type 4.5 and press EXE. The -value appears at the bottom of the screen. Press EXIT to finish. when for the exponential model, to 1 decimal place. |
X=4.5 Y=59.0330102 |
A sinusoidal model | |
Enter the times in List 1 and the depths in List 2: List 1 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24; List 2 3.1, 5.6, 7, 6.4, 3.9, 1.6, 1.3, 3.4, 5.9, 7.1, 6.2, 3.8, 1.9. Press F2 (CALC), F3 (REG), F6 and F4 (Sin). The screen gives a, b, c, d and MSe and no r². , with the coefficients to 3 significant figures. |
Sinusoidal Reg a=2.9173012 b=0.4526533 c=-0.3835108 d=4.2119403 MSe=... |
Drawing a sine regression graph switches the angle unit to radians, but the coefficients are in radians whatever the setting. Use radian mode whenever you evaluate the equation yourself. |
|
Check the model has the form your exam question expects: use abx, not aebx, for y = ab^x. A sinusoidal regression gives no r².
How to fit a model on a Casio fx-CG100
Enter the data | |
From HOME open Statistics; the List Editor tab is showing. |
|
Fill List 1 with the inputs, EXE after each value: 0, 1, 2, 3, 4, 5, 6, 7. Then fill List 2 with the outputs the same way: 12, 18, 25, 36, 49, 70, 98, 141. The lists need equal lengths. |
List1 = 0, 1, 2, 3, 4, 5, 6, 7 List2 = 12, 18, 25, 36, 49, 70, 98, 141 |
Choose the model and run it | |
Press the next-tab key to reach the Setup tab. Highlight the first line and press OK, then choose Calculate Summary Statistics and press OK. |
|
Highlight line 2 and press OK, then choose the model: Quadratic Regression, Cubic Regression, Exp Regression (a·b^x), Power Regression or Sinusoidal Regression. For the example choose Exp Regression (a·b^x) for . Check that 2Var XList is List 1 and 2Var YList is List 2. |
Calculate Summary Statistics Exp Regression(a·b^x) 2Var XList: List1 2Var YList: List2 |
Press the next-tab key, or highlight Execute and press OK. The Results tab shows a, b, r, r² and MSe. and , with the coefficients to 3 significant figures. |
Exp Regression(a·b^x) a=12.4068849 b=1.4142966 r=0.9997204 r²=0.9994408 MSe=... |
Compare two models with R² | |
Go back to the Setup tab, change line 2 to another model such as Quadratic Regression, and execute again. The Results tab shows r² for the quadratic, cubic, exponential and power models, but not for the sinusoidal model. Compare the values, then look at the graph before you decide. and . |
Quadratic Regression a=2.8392857 b=-2.6011905 c=15.5416667 r²=0.9936753 MSe=... |
Predict a value from the model | |
Go back to the Setup tab and choose Draw Statistics Graph on the first line. With Graph1 as a scatter plot of List 1 against List 2, press the next-tab key to draw it, then the next-tab key again and choose the regression model from the menu, and the next-tab key once more to draw the regression graph. |
|
Press TOOLS and choose Graph Solve then y-Cal. Type 4.5 and press OK. The coordinates appear at the bottom of the window. when for the exponential model, to 1 decimal place. |
x=4.5 y=59.0330102 |
A sinusoidal model | |
Enter the times in List 1 and the depths in List 2: List 1 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24; List 2 3.1, 5.6, 7, 6.4, 3.9, 1.6, 1.3, 3.4, 5.9, 7.1, 6.2, 3.8, 1.9. Choose Sinusoidal Regression on line 2 of the Setup tab and execute. The Results tab shows a, b, c, d and MSe and no r². , with the coefficients to 3 significant figures. |
Sinusoidal Regression a=2.9173012 b=0.4526533 c=-0.3835108 d=4.2119403 MSe=... |
Choose the form with b to the power x for y = ab^x, not the one with e. The sinusoidal regression gives no r².
How to fit a model on a TI-Nspire CX
Enter the data | |
From home open Lists & Spreadsheet. Name the lists in the top cells: |
|
Put the inputs down column A, enter after each value: 0, 1, 2, 3, 4, 5, 6, 7. Put the outputs down column B: 12, 18, 25, 36, 49, 70, 98, 141. The columns need equal lengths. |
x: 0, 1, 2, 3, 4, 5, 6, 7 y: 12, 18, 25, 36, 49, 70, 98, 141 |
Choose the model and run it | |
Press menu → Statistics → Stat Calculations. Choose the model: Quadratic Regression, Cubic Regression, Exponential Regression (), Power Regression or Sinusoidal Regression. |
|
For the example choose Exponential Regression. In the box, set X List to and , with the coefficients to 3 significant figures. |
a=12.4068849 b=1.4142966 r²=0.9994408 r=0.9997204 |
Compare two models with R² | |
Run another model on the same columns: menu → Statistics → Stat Calculations → Quadratic Regression, with the same X List and Y List. The results include R² for the quadratic, cubic, exponential and power models. Compare the values, then look at the graph before you decide. and . |
a=2.8392857 b=-2.6011905 c=15.5416667 R²=0.9936753 |
Predict a value from the model | |
In the regression box leave Save RegEqn To as when for the exponential model, to 1 decimal place. |
f1(4.5) 59.0330102 |
A sinusoidal model | |
Enter the times in column A and the depths in column B: column A 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24; column B 3.1, 5.6, 7, 6.4, 3.9, 1.6, 1.3, 3.4, 5.9, 7.1, 6.2, 3.8, 1.9. Choose Sinusoidal Regression. The box has X List and Y List and also Iterations and Period: leave Period at its default for equally spaced times, or type a period guess. The results are a, b, c and d, with no R². , with the coefficients to 3 significant figures. |
a=2.9173012 b=0.4526533 c=-0.3835108 d=4.2119403 |
The angle is always in radians, so put the document in radian mode before you evaluate |
|
Give the lists names such as x and y and keep Save RegEqn To as f1. The results are linked to the data, so if you correct a value in the columns the model updates.
Sources. Key presses follow the manufacturers' documentation: the TI-84 Plus CE eGuide, the TI-Nspire eGuide, the Casio fx-CG50 Software User's Guide (version 3.30) and the Casio fx-CG100 online manual.
What to write in the exam
- Write the equation of the model with its parameters to 3 significant figures, for example , and note that technology found it. The key presses do not belong in the working.
- Match the letters in the question. If the data are called and , give the model as in terms of .
- When R² is asked for, quote it and interpret it: a value near 1 means the curve follows the data closely. A higher R² does not by itself make a model suitable, because the shape must also fit the context.
- Stay within the data when you estimate. A value of between the smallest and largest in the table gives a trustworthy estimate (interpolation); one beyond them is extrapolation and needs a comment.
- Calculators are allowed on every AI HL paper (check your syllabus), so regression is available throughout.
Choosing a model
Plot the data first (see making a scatter diagram) and let the shape and the context guide you.
- A curve with a single turning point suggests a quadratic, and one with two turning points suggests a cubic.
- Growth or decay by a constant percentage each step suggests an exponential model .
- A relationship where doubling multiplies by a fixed factor suggests a power model .
- A repeating, wave-like pattern suggests a sinusoidal model .
Comparing models with R²
R² lies between 0 and 1, and the closer to 1 the more of the variation in the model accounts for. For the exponential and power models the calculators work R² out on the straightened data ( against , or against ), so compare it with the other models as a guide and check the fit on the graph as well. A cubic fits any data at least as well as a quadratic, so a higher R² alone does not make it the better choice. The sinusoidal regression does not give R². To find when a model reaches a given value, see solving an equation numerically.
Error messages and what to do
| Calculator | Message | Usual cause | What to try |
|---|---|---|---|
| TI-84 Plus CE | ERR:DOMAIN | An exponential or power regression was run on data with a negative y-value (or a power regression with a negative x-value), which has no logarithm. | Check the data, and choose a model that suits it. |
| TI-84 Plus CE | ERR:DIMENSION MISMATCH | L1 and L2 do not have the same number of values. | Open the list editor with STAT then 1:Edit and make both lists the same length. |
| TI-84 Plus CE | ERR:SINGULAR MATRIX | SinReg or a polynomial regression could not find a solution, for example with too few data points. | Check the data, add points if you can, and for SinReg type a period guess. |
| TI-84 Plus CE | ERR:STAT | The lists are not suitable for the calculation, for example fewer than two data points. | Check each list holds all your values. |
| Casio fx-CG50 | Dimension ERROR | The two lists do not have the same number of values. | Make List 1 and List 2 the same length. |
| Casio fx-CG50 | Ma ERROR | The calculation went outside the allowed range, for example an exponential or power regression on data with a zero or negative value. | Check the data, and choose a model that suits it. |
| Casio fx-CG100 | Dimension ERROR | The two lists do not have the same number of values. | Make List 1 and List 2 the same length. |
| Casio fx-CG100 | Math ERROR | The calculation went outside the allowed range, for example an exponential or power regression on data with a zero or negative value. | Check the data, and choose a model that suits it. |
| TI-Nspire CX | Dimension mismatch | The X List and Y List do not have the same number of values. | Make columns A and B the same length. |
| TI-Nspire CX | Domain error | An exponential or power regression was run on data with a zero or negative value. | Check the data, and choose a model that suits it. |
Common mistakes
- Putting the variables in the wrong lists. The input goes in the first list and the output in the second. Reversing them fits a different curve.
- Missing R² on the TI-84 Plus CE. It is hidden while STAT DIAGNOSTICS is off, so switch it on in the mode screen and run the model again.
- The wrong exponential form. For choose the form with , not the one with , or the parameters will not match the question.
- Degrees instead of radians for a sinusoidal model. The calculator gives and in radians. Use radian mode when you evaluate the equation yourself.
- Estimating from rounded parameters. Evaluate the stored equation, or keep every digit of the parameters, and round once at the end.
- Trusting a model far outside the data. Even with R² close to 1, an estimate well beyond the recorded values may be wrong.
Try it yourself
Real questions from the IB Maths Help question bank. Type your answer and press Mark it: the worked solution and mark scheme then appear. Use your calculator to fit the model.
Practise with your calculator
Choose your course to see the practice sets that use this skill. You can also try practice mode or browse every GDC guide.
Quick questions
How do I fit an exponential or quadratic model on a calculator?
Enter the x-values in one list and the y-values in another, then run the regression for the model: ExpReg or QuadReg on the TI-84 Plus CE, CALC then REG then Exp or X squared on the Casio fx-CG50, a Calculate Summary Statistics regression on the fx-CG100, or Statistics then Stat Calculations on the TI-Nspire. The calculator gives the parameters of the equation.
Why does my TI-84 not show R squared?
R squared only appears when STAT DIAGNOSTICS is on. Turn it on in the mode screen, then run the regression again. The sinusoidal regression never shows it.
Which model should I choose?
Look at the scatter diagram and the context. Curves with turning points suggest quadratic or cubic, constant percentage change suggests exponential, a power relationship suggests power, and a repeating pattern suggests sinusoidal. Compare R squared for the candidates, but also check the model makes sense.
Why are my sinusoidal values different from the answer in the book?
A sinusoidal curve can be written in more than one equivalent form, and the calculator finds its answer by repeated approximation. Compare the graph of your equation with the data, and check the calculator is giving the angle in radians.