TI TI84 Calculator Editorial Team

How to Do Linear Regression on TI-84 — Complete Step-by-Step Guide (2026)

Linear regression on a TI-84 involves four main steps: enter your x-values in L1 and y-values in L2, run LinReg(ax+b) from the STAT CALC menu, read the slope (a), intercept (b), and correlation coefficient (r) from the output, and optionally graph the regression line over your scatter plot. The whole process takes under two minutes once you know the path. This guide walks through each step with a worked example, explains what every output value means, and covers the common mistakes that cause wrong answers or missing r values.

⚡ Quick Answer

Enter x-values in L1 and y-values in L2 via STAT → Edit. Then press STAT → CALC → 4: LinReg(ax+b) → set Xlist: L1, Ylist: L2, Store RegEQ: Y₁ → Calculate. Read a (slope), b (y-intercept), r (correlation), and (coefficient of determination). Press GRAPH to see the line on your scatter plot.

✅ Key Takeaways
  • The regression equation is ŷ = ax + b where a is the slope and b is the y-intercept.
  • r measures strength and direction of the linear relationship (−1 to +1).
  • tells you what percentage of variation in y is explained by x.
  • If r and r² are not showing, you need to enable DiagnosticOn first.
  • Store the equation in Y₁ to graph the regression line with one button press.
  • Always check your list data before running regression — leftover values from a previous session will corrupt results.
  • No TI-84 handy? The free online TI-84 simulator supports full regression workflows.

Before You Start — Enable DiagnosticOn

By default, the TI-84 hides the correlation coefficient (r) and coefficient of determination (r²) from the regression output. You need to turn on DiagnosticOn once — it stays enabled even after the calculator is turned off.

💡 Only needs to be done once DiagnosticOn persists across sessions and battery changes. You'll never need to re-enable it unless you perform an All Memory reset. If you've recently reset your calculator, see the TI-84 reset guide for what that clears.

Step 1 — Enter Data into L1 and L2

Regression requires two matched lists: L1 for x-values (the independent variable) and L2 for y-values (the dependent variable). Every x-value must have a corresponding y-value at the same row position.

⚠️ List length mismatch = ERR: DIM MISMATCH If L1 has 5 values and L2 has 4, the calculator throws a dimension mismatch error when you try to run LinReg. Always confirm both lists have the same number of entries before proceeding.

Step 2 — Create a Scatter Plot (Optional but Useful)

Viewing a scatter plot before running regression lets you visually confirm the data looks linear — not curved or U-shaped — which makes the linear model appropriate. It also lets you spot any obvious outliers before they influence the equation.

📌 Turn off STAT PLOT before regular graphing After regression, if you return to graphing equations and see unexpected dots on screen, STAT PLOT 1 is still on. Press 2ndY= → Plot1 → Off to disable it. Full graphing tips are in the TI-84 graphing guide.

Step 3 — Run LinReg(ax+b)

With data in your lists, you're ready to run the linear regression calculation. The function is LinReg(ax+b) — option 4 in the STAT CALC menu.

Step 4 — Reading and Interpreting the Output

The LinReg output screen shows four values. Here's what each one means and how to use it:

LinReg(ax+b) Output y = ax+b
a = 2.01 ← slope
b = 1.18 ← y-intercept
r² = 0.9971 ← coefficient of determination
r = 0.9985 ← correlation coefficient
SymbolNameWhat It Tells You
a Slope For each 1-unit increase in x, y increases by a units. Negative a = inverse relationship.
b Y-intercept Predicted value of y when x = 0. May not have real-world meaning depending on context.
Coefficient of Determination Proportion of variation in y explained by x. r² = 0.997 means 99.7% of y's variation is explained by the linear model — an excellent fit.
r Correlation Coefficient Strength and direction of the linear relationship. Ranges from −1 to +1. Values near ±1 indicate strong linearity; near 0 indicates weak or no linear relationship.

How to Write the Regression Equation

Using the sample output above, the regression equation is:

Regression Equation
ŷ = 2.01x + 1.18
a = 2.01 (slope)  |  b = 1.18 (y-intercept)

To predict y for a specific x-value, substitute the x into the equation. For x = 6: ŷ = 2.01(6) + 1.18 = 13.24. You can also use the TRACE function on the graph to read predicted values directly — covered in the next section.

Step 5 — Graphing the Regression Line

If you stored the equation in Y₁ during step 3, graphing the line alongside your scatter plot takes two button presses.

Full Worked Example with Real Data

Let's apply every step to a concrete problem.

Scenario: A student records hours studied and exam scores for 6 classmates.

Hours Studied (x)Exam Score (y)
152
258
365
471
578
685

Entering Data

Running LinReg

Results

LinReg Output — Hours vs Score y = ax+b
a = 6.6286 ← score increases ~6.6 pts per hour
b = 45.667 ← predicted score at 0 hours
r² = 0.9983 ← 99.8% of score variation explained
r = 0.9991 ← very strong positive correlation

Regression equation: ŷ = 6.63x + 45.67

Interpretation: Each additional hour of study predicts about 6.6 more points on the exam. The correlation of 0.999 indicates an almost perfectly linear relationship in this data set.

Prediction: For a student who studies 7 hours: ŷ = 6.63(7) + 45.67 = 92.1

📌 Connecting to standard deviation The correlation coefficient r can also be expressed as r = (1/n−1) × Σ[(x−x̄)/Sₓ × (y−ȳ)/Sᵧ]. To review how Sx is calculated and what the 1-Var Stats output means, see the standard deviation guide.

Other Regression Types Available on TI-84

Linear regression is the most common, but the TI-84's STAT CALC menu includes several other regression models. Use these when a scatter plot suggests a non-linear pattern.

LinReg(ax+b)
Linear — use first
QuadReg
Quadratic / U-shaped
CubicReg
Cubic / S-shaped curve
ExpReg
Exponential growth/decay
PwrReg
Power functions
LnReg
Logarithmic
SinReg
Sinusoidal / periodic
LinReg(a+bx)
Linear — alternate form

All regression types follow the same workflow: enter data in lists, select the appropriate function from STAT → CALC, specify L1 and L2, optionally store in Y₁, and press Calculate. The output format varies slightly but always includes the equation coefficients.

💡 Which regression model to choose? Start by graphing your scatter plot. If the points form a straight band, use LinReg. If they curve upward with increasing speed, try ExpReg or PwrReg. If they form a U or inverted-U, QuadReg. The model with the highest r² value (closest to 1) is usually the best fit.

Common Mistakes and How to Fix Them

ProblemCauseFix
r and r² not showing in output DiagnosticOn not enabled Press 2nd → CATALOG → scroll to DiagnosticOn → ENTERENTER. Now re-run LinReg.
ERR: DIM MISMATCH L1 and L2 have different numbers of entries Go to STAT → Edit and count rows. Delete extra values or add missing ones so both lists match.
Wrong regression equation — impossible slope Old data still in L1 or L2 from previous session Clear lists before entering new data: highlight the list header in STAT Edit → CLEARENTER. Or see the reset guide for ClrAllLists.
Regression line not appearing on graph Equation was not stored in Y₁, or Y₁ is disabled Re-run LinReg and include Y₁ as the Store RegEQ destination. Check that Y₁'s = sign is highlighted in the Y= menu.
Scatter plot not visible with regression line STAT PLOT is turned off Press 2ndY= → Plot1 → On. Then press ZOOM → ZoomStat to redraw.
r is negative when relationship looks positive X and Y lists entered in wrong order (swapped) Check that your x-values are in L1 and y-values are in L2 — not the other way around. The sign of r matches the sign of the slope a.
TRACE not showing ŷ values on the regression line Cursor is on the scatter plot dots, not the Y₁ line While in TRACE mode, press ↑ or ↓ to switch between the scatter plot and the stored regression equation. The top-left shows which function is active.

Practice Linear Regression Right Now — Free

Open the free TI-84 online simulator, enter your data in L1 and L2, and run LinReg — no physical calculator needed.

Open Free TI-84 Simulator →

Linear regression sits at the intersection of several statistics skills. These guides cover the techniques that connect directly to it:

Frequently Asked Questions

How do I do linear regression on a TI-84 calculator?

Enter x-values in L1 and y-values in L2 via STAT → Edit. Then press STAT → CALC → 4: LinReg(ax+b). Set Xlist to L1 and Ylist to L2 — optionally store the equation in Y₁. Press Calculate. The output shows a (slope), b (y-intercept), , and r.

Why is r not showing in my TI-84 regression output?

The TI-84 hides r and r² by default. You need to enable DiagnosticOn: press 2nd0 (CATALOG) → scroll to DiagnosticOn → press ENTER twice. The screen confirms with "Done." Now re-run your LinReg and r will appear. This setting persists until an All Memory reset.

What is the difference between a and b in TI-84 LinReg output?

In the equation ŷ = ax + b, a is the slope — it tells you how much y changes for each 1-unit increase in x. b is the y-intercept — the predicted value of y when x equals zero. For example, if a = 3.5 and b = 2.0, the equation is ŷ = 3.5x + 2.0, meaning each additional unit of x adds 3.5 to the predicted y.

How do I graph the regression line on TI-84?

When running LinReg, store the equation in Y₁ by selecting it in the Store RegEQ field (or append Y₁ to the command on the home screen). Then make sure STAT PLOT 1 is enabled with L1/L2 selected. Press ZOOM9: ZoomStat — the scatter plot and regression line both appear. Use TRACE and arrow keys to read predicted values along the line.

What does r² mean in linear regression on TI-84?

r² (the coefficient of determination) tells you what proportion of the variation in your y-values is explained by the linear relationship with x. An r² of 0.95 means 95% of the variation in y is accounted for by the model — only 5% is unexplained. Values closer to 1 indicate a better-fitting linear model. An r² below 0.5 generally suggests a weak linear fit.

What is ERR: DIM MISMATCH in TI-84 regression?

This error appears when L1 and L2 have different numbers of entries. Every x-value in L1 needs a matching y-value in L2 at the same row position. To fix it: go to STAT → Edit and count the rows in each list. Add the missing value or remove the extra one. If old data is causing confusion, clear both lists completely by highlighting each list header and pressing CLEARENTER.

Can I do linear regression on the TI-84 online simulator?

Yes. The free TI-84 simulator at ti84calculato.com supports the full STAT menu including LinReg(ax+b), list editing, and scatter plots via STAT PLOT. Every step in this guide works identically on the online version — useful for practice or for verifying results when you don't have a physical calculator available.

How do I predict a y-value from my regression equation on TI-84?

There are two methods. First, substitute the x-value directly into the equation on the home screen: type a × x + b using your actual values from the LinReg output. Second (easier): after storing the equation in Y₁ and graphing, press TRACE, make sure the cursor is on the Y₁ line, type your x-value, and press ENTER. The calculator jumps to that point and displays the predicted ŷ at the bottom of the screen.

Linear Regression in Five Steps

The complete regression workflow on a TI-84 follows a consistent path: enable DiagnosticOn once, enter paired data in L1 and L2, run LinReg(ax+b) from STAT → CALC, read the slope and intercept from the output, and graph the line by storing the equation in Y₁. The r and r² values tell you immediately how well the linear model fits — values above 0.95 generally indicate a strong relationship worth using for predictions.

Regression connects naturally to other statistics skills on the TI-84. The same list workflow you use here applies to 1-Var Stats for standard deviation, and the correlation coefficient r links directly to Z-scores and normal distribution calculations. For the full collection of TI-84 tutorials covering every exam-relevant function, visit the Guides & Tutorials section.