Question about Texas Instruments TI-89 Calculator

It's the "Exact/Approx" setting under the [MODE] key. I used to have it on "Exact" and got the domain error issue, but then changed it to "Auto" and it works. IMO, this is a really weird quirk that TI should have ironed out before releasing the TI-89.

Posted on Jan 08, 2010

I changed the mode and that solved the problem.

Posted on Apr 03, 2008

It doesnt work if you type factions into your lists. place your cusor over the name of your list, i.e. list1 (make sure its highlighted), press the diamond key and then press enter. This should convert all the fractions to decimals. Doing this made the domain error for LinReg(ax+b) go away for me. Good luck!

Posted on Dec 03, 2009

What do you mean? Which part of the mode did you switch?because I'm having the same problem.

Posted on Sep 03, 2008

DOMAIN ERROR:

- You specified an argument to a function or

instruction outside the valid range. This error is

not returned during graphing. The TI-84 Plus

allows for undefined values on a graph.

- You attempted a logarithmic or power

regression with a LX or an exponential or power

regression with a LY.

- You attempted to compute GPrn( or GInt( with

pmt2 < pmt1.

- You specified an argument to a function or

instruction outside the valid range. This error is

not returned during graphing. The TI-84 Plus

allows for undefined values on a graph.

- You attempted a logarithmic or power

regression with a LX or an exponential or power

regression with a LY.

- You attempted to compute GPrn( or GInt( with

pmt2 < pmt1.

Mar 16, 2015 | Texas Instruments TI-84 Plus Calculator

This means that you're invoking some function with an argument out of the function's domain. For example, the arcsine function requires an argument between -1 and +1, inclusive. If you try to take the arcsine of 2, you'll get this error.

If you require further assistance, please reply to this post and specify exactly what you're trying to do when you get this error.

If you require further assistance, please reply to this post and specify exactly what you're trying to do when you get this error.

Mar 07, 2014 | Texas Instruments TI-83 Plus Calculator

Here is a copy on an earlier post I made detailing how to do a regression on a TI8x Plus families of calculators. In the post I took a quadratic regression, but that should not be a problem for you to use the same techniques. Just, when you reach the point where you select the regression model, select SinReg.

If you follow the instructions you should be able to get the scatter plot and the best-fit curve on the same graph. If you do get an error check your data and blame not the method I will show you because it is sound.

Here it goes

Your question is a very interesting one and I believe people will benefit from a complete solution to it. So I will provide a solution for a quadratic regression.

The points I will represent are those on a parabola

L1= {0,1,2,3,4,5,6,7,8,9,10}

L2={0,1,4,9,16,25,36,49,64,81,100}

As you can see, I am cheating (no experimental points will fit a perfect parabola) but this will have no incidence on what I am going to show you.

1. PREPARE THE GRAPH FORMAT (you will see why latter)

With the calculator turned On, press [2nd][ZOOM] to access the (FORMAT) screen. Notice that RectGC is highlighted. It means that the coordinates are rectangular. IF the POLARGC had been highlighted and you had used the TRACE function the values of the coordinates of the points being traced would be in the form (R, theta), in other words polar coordinates.

2. ENTER THE DATA

Press STAT, then under the EDIT menu press [1:Edit]

to open the list editing screen. Enter the elements of the two lists

After you enter the data in L1 and L1, Press [2nd][Y=] to open the (STAT PLOT)

3. CONFIGURE THE STAT PLOT TYPE

Initially the Plot1 may be OFF. Make sure at the beginning that all Stat Plots are OFF. If necessary press [4:PlotsOff] execute the command in Main calculator screen and press [2nd][Y=1] to get back at the stat plot screen configuration below

Press 1:Plot1 Off to open the P1 configuration

The cursor will be blinking on On. If not use cursor to highlight ON and press [ENTER]. The cursor will move to second line (TYPE), on first icon. Since you want to do a scatter PLOT that is the one to choose. Highlight it and press ENTER then move cursor with Down arrow to move to Xlist

If your data is in L1 and L2 Xlist and Ylist will be already selected. If you have entered your data in some other lists, enter the proper lists (L3,L4, etc.)

DRAWING THE SCATTER PLOT

Now you can press the [GRAPH] function key. Here is what you get

PERFORMING A REGRESSION

Because I chose it so, we will perform a quadratic regression QuadReg

To do so you press [STAT] and use right arrow to select the CALC(ulate) Menu. Type 5:QuadReg

Command echoes on Main screen as QuadReg with blinking cursor as above on the right.

If you press [ENTER] the calculator will assign L1 to X and L2 to Y by default. If that does not correspond to the problem, you had better change the names of the lists used by entering then explicitly on the command line. Just press [2nd][L1] then [2nd] [L2] to get the following screen

Pressing [ENTER] to execute the command will get you the regression equation and its coefficients but you will not be able to have it graphed on the scatter plot

The coefficient/factor of determination R2 is displayed if you ran the DiagnosticOn command from the catalog, or as in the newer version of the OS (version 2.53MP) you press [MODE], scroll down to get to the second page of options and selected Stat Diagnostics ON

Thus to be able to draw the regression curve on the scatter plot, you must save it to a function variable, Y1 for example. You do it with the following command line

After you press ENTER you obtain the equation, a,b,c, and R2. By pressing the [GRAPH] function key you get the raw data and the regression on the same screen.

As you can see both are represented

TRACING THE REGRESSION CURVE

Press Trace and see what you get

As you can see the trace defaults to the scatter plot P1, L1,L2 (see the top left). If you stay on that graph you will on ly trace the actual data points (raw data). The cursor will jump from one point to the next.

To trace the regression curve (the second curve) you must move cursor on it. To do so press the Up and Down arrows. Left and right arrows allow cursor to move along the same curve.

Pressing Down Arrow you move to the regression curve. Notice the equation on the top left corner. The cursor can be displaced anywhere (depending on the resolution of the graph.

If you want Y for a particular value of X, just enter the value of X

Enter an X value. How about 5.55555?

If as I have suggested above, you copied the regression equation to Y1 variable, you can calculate Y-values for any X in the graph domain.

In Main calculator screen type the numerical value of X, say 4.777756, and store it in X-variable

4.777756 [STO>] [X,T,theta,n] [ENTER]

Calculator answers Done

Press [VARS][>][Y-VARS][1:Function...][1:Y1] [ENTER]

The calculator computes the value of Y1 for the entered value of X.

You can also generate a whole table, but you will have less control over the points where Y1 will be calculated.

If you follow the instructions you should be able to get the scatter plot and the best-fit curve on the same graph. If you do get an error check your data and blame not the method I will show you because it is sound.

Here it goes

Your question is a very interesting one and I believe people will benefit from a complete solution to it. So I will provide a solution for a quadratic regression.

The points I will represent are those on a parabola

L1= {0,1,2,3,4,5,6,7,8,9,10}

L2={0,1,4,9,16,25,36,49,64,81,100}

As you can see, I am cheating (no experimental points will fit a perfect parabola) but this will have no incidence on what I am going to show you.

1. PREPARE THE GRAPH FORMAT (you will see why latter)

With the calculator turned On, press [2nd][ZOOM] to access the (FORMAT) screen. Notice that RectGC is highlighted. It means that the coordinates are rectangular. IF the POLARGC had been highlighted and you had used the TRACE function the values of the coordinates of the points being traced would be in the form (R, theta), in other words polar coordinates.

2. ENTER THE DATA

Press STAT, then under the EDIT menu press [1:Edit]

to open the list editing screen. Enter the elements of the two lists

After you enter the data in L1 and L1, Press [2nd][Y=] to open the (STAT PLOT)

3. CONFIGURE THE STAT PLOT TYPE

Initially the Plot1 may be OFF. Make sure at the beginning that all Stat Plots are OFF. If necessary press [4:PlotsOff] execute the command in Main calculator screen and press [2nd][Y=1] to get back at the stat plot screen configuration below

Press 1:Plot1 Off to open the P1 configuration

The cursor will be blinking on On. If not use cursor to highlight ON and press [ENTER]. The cursor will move to second line (TYPE), on first icon. Since you want to do a scatter PLOT that is the one to choose. Highlight it and press ENTER then move cursor with Down arrow to move to Xlist

If your data is in L1 and L2 Xlist and Ylist will be already selected. If you have entered your data in some other lists, enter the proper lists (L3,L4, etc.)

DRAWING THE SCATTER PLOT

Now you can press the [GRAPH] function key. Here is what you get

PERFORMING A REGRESSION

Because I chose it so, we will perform a quadratic regression QuadReg

To do so you press [STAT] and use right arrow to select the CALC(ulate) Menu. Type 5:QuadReg

Command echoes on Main screen as QuadReg with blinking cursor as above on the right.

If you press [ENTER] the calculator will assign L1 to X and L2 to Y by default. If that does not correspond to the problem, you had better change the names of the lists used by entering then explicitly on the command line. Just press [2nd][L1] then [2nd] [L2] to get the following screen

Pressing [ENTER] to execute the command will get you the regression equation and its coefficients but you will not be able to have it graphed on the scatter plot

The coefficient/factor of determination R2 is displayed if you ran the DiagnosticOn command from the catalog, or as in the newer version of the OS (version 2.53MP) you press [MODE], scroll down to get to the second page of options and selected Stat Diagnostics ON

Thus to be able to draw the regression curve on the scatter plot, you must save it to a function variable, Y1 for example. You do it with the following command line

After you press ENTER you obtain the equation, a,b,c, and R2. By pressing the [GRAPH] function key you get the raw data and the regression on the same screen.

As you can see both are represented

TRACING THE REGRESSION CURVE

Press Trace and see what you get

As you can see the trace defaults to the scatter plot P1, L1,L2 (see the top left). If you stay on that graph you will on ly trace the actual data points (raw data). The cursor will jump from one point to the next.

To trace the regression curve (the second curve) you must move cursor on it. To do so press the Up and Down arrows. Left and right arrows allow cursor to move along the same curve.

Pressing Down Arrow you move to the regression curve. Notice the equation on the top left corner. The cursor can be displaced anywhere (depending on the resolution of the graph.

If you want Y for a particular value of X, just enter the value of X

Enter an X value. How about 5.55555?

If as I have suggested above, you copied the regression equation to Y1 variable, you can calculate Y-values for any X in the graph domain.

In Main calculator screen type the numerical value of X, say 4.777756, and store it in X-variable

4.777756 [STO>] [X,T,theta,n] [ENTER]

Calculator answers Done

Press [VARS][>][Y-VARS][1:Function...][1:Y1] [ENTER]

The calculator computes the value of Y1 for the entered value of X.

You can also generate a whole table, but you will have less control over the points where Y1 will be calculated.

Feb 13, 2011 | Texas Instruments TI-84 Plus Calculator

This means that the two lists containing your data do not contain the same number of points. If the data are not in L1 and L2, make sure you specify the two lists involved to the regression command.

Oct 21, 2010 | Texas Instruments TI-84 Plus Calculator

Start by creating the data lists.

CREATE THE X-LIST

Using Tab to move from one line to the next, configure the calculator as seen on the next screen.

And the last line is at the bottom of the next screen capture

The statistical results will hence be displayed in the C[] column.

Use TAb to move focus to OK, and press ENTER.

Here they are.

You can recognize each statistical results, RegEq is y=mx+b. You also have m, b, r^2, and r. The residuals appear on the next row. By highlighting the cell with the {0.,0...0} you can display the hidden values

Here you have all you asked for. But you can go further by drawing the curve.

The regression equation is stored in function f1, so you have to draw the f1 function. .

DRAWING THE REGRESSION CURVE

Press the HOME button

Highlight and select [2:Graphs and Geometry]

At the bottom of the screen there is a command line with f2(x)=

Use the UpArrow to display the f1(x)= equation

Press ENTER to draw the function f1

I think I have to stop here. There is enough material to get you started.

I left for you the part where you draw the data points. read the user manual.

CREATE THE X-LIST

- Press Home.
- Use navigation pad to Highlight [3: Lists and Spreadsheets]
- Press Enter
- A list opens with cursor in Cell A1.
- Use UpArrow to move the cursor to the header of the list and give it a name, say X.
- Press Enter. Cursor moves to Cell A1
- Enter an X-value in each cell and press ENTER to move to next cell:
- Type 1, and press ENTER. Say Yes to overwrite message by pressing OK.
- Enter 2,3,4,5,6,7,8,9,10 (just a bogus sample).

- After typing 10 in cell A10, press ENTER to validate.
- Use the Navigation pad to move cursor to the head of the B column.
- Give it a name , say Y.
- Press ENTER, Cursor moves to cell B1.
- As for X-list, enter the Y values. Here is a sample (-2, -1, 0, 1,2,3,4,5,6,7). After you type value in B1, accept the overwrite message by pressing OK. Type the other values.
- Make sure that the X-List and Y-List have the same number of elements, otherwise you will have a dimension mismatch error.

- While the lidt is still displyed, prress the Menu key
- Highlight [4:Statistics]
- Use Right Arrow to unfold submenu list and select [1:Stat Calculations]
- Use right Arrow to unfold the submenu list and
- Select [3:Linear regression (mx+b)] or [4:linear regression (ax+b)]

Using Tab to move from one line to the next, configure the calculator as seen on the next screen.

And the last line is at the bottom of the next screen capture

The statistical results will hence be displayed in the C[] column.

Use TAb to move focus to OK, and press ENTER.

Here they are.

You can recognize each statistical results, RegEq is y=mx+b. You also have m, b, r^2, and r. The residuals appear on the next row. By highlighting the cell with the {0.,0...0} you can display the hidden values

Here you have all you asked for. But you can go further by drawing the curve.

The regression equation is stored in function f1, so you have to draw the f1 function. .

DRAWING THE REGRESSION CURVE

Press the HOME button

Highlight and select [2:Graphs and Geometry]

At the bottom of the screen there is a command line with f2(x)=

Use the UpArrow to display the f1(x)= equation

Press ENTER to draw the function f1

I think I have to stop here. There is enough material to get you started.

I left for you the part where you draw the data points. read the user manual.

Jan 28, 2010 | Texas Instruments TI-Nspire Graphic...

Hello,

The correlation coefficient does not have meaning (is not defined) for most regression models. Check you theory to make sure it exits for the sine regression.

An easy way to verify that, is to use the Command DiagnosticOn, accessible through the catalog [2nd][0] (CATALOG) . Press [X^-1] (D) to jump to first command that starts with D. Scroll down to DiagnosticOn, select it and press [ENTER]. Calculator replies Done.

Do your sine regression again. If the correlation coefficient has meaning for this type of regression, you will find it with the other statistical variables. You may have to scroll down the statistical variable list to verify the whole contents of the list.

Hope it helps.

Thank you for rating the solution.

The correlation coefficient does not have meaning (is not defined) for most regression models. Check you theory to make sure it exits for the sine regression.

An easy way to verify that, is to use the Command DiagnosticOn, accessible through the catalog [2nd][0] (CATALOG) . Press [X^-1] (D) to jump to first command that starts with D. Scroll down to DiagnosticOn, select it and press [ENTER]. Calculator replies Done.

Do your sine regression again. If the correlation coefficient has meaning for this type of regression, you will find it with the other statistical variables. You may have to scroll down the statistical variable list to verify the whole contents of the list.

Hope it helps.

Thank you for rating the solution.

Dec 13, 2009 | Texas Instruments TI-84 Plus Silver...

Hello,

1.Press [MODE], scroll to set**Func**tion graphing mode.

2.Press [STAT] [5:SetUpEditor], command is pasted in home screen. Press [ENTER]. Screen displays Done. This clears old values in lists

3. Press [STAT][1:Edit] to enter data.

Press [ENTER] after each number. Cursor moves to next row. After finishing the data of Listt L1, use arrow right and UP to the head of list L2. Enter the values in list L2. Press Enter after last element.

4. Press [Y=] to display Y= editor. Clear function Y1 and all functions or remove highlight from the = in Y2, Y3, etc.

5.Press [2nd][STATPLOT]1 to select 1:Plot1

The blinking cursor on ON means graph will be drawn. Type chosen is Scatter plot. Mark chosen is a dot.

Press [ZOOM][9:ZoomStat]

If you think that the curve look like a straight line you can try to do a linear regression.

Press [STAT] Arrow right to highlight [CALC]

Select [4: LinReg(ax+b)]

LinReg(ax+b) echoes on run screen. Press [2nd] L1 Press comma, Press [2nd] L2 press comma Press [VARS] Arrow right to select [Y:VARs][1:Function] [ENTER][1:Y1]

ENTER

That is the equation for the linear regression. As you see eqaution has been saved in Y1. Press [GRAPH] to get the data and the regression curve drawn on same graph.

Hope it helps.

1.Press [MODE], scroll to set

2.Press [STAT] [5:SetUpEditor], command is pasted in home screen. Press [ENTER]. Screen displays Done. This clears old values in lists

3. Press [STAT][1:Edit] to enter data.

Press [ENTER] after each number. Cursor moves to next row. After finishing the data of Listt L1, use arrow right and UP to the head of list L2. Enter the values in list L2. Press Enter after last element.

4. Press [Y=] to display Y= editor. Clear function Y1 and all functions or remove highlight from the = in Y2, Y3, etc.

5.Press [2nd][STATPLOT]1 to select 1:Plot1

The blinking cursor on ON means graph will be drawn. Type chosen is Scatter plot. Mark chosen is a dot.

Press [ZOOM][9:ZoomStat]

If you think that the curve look like a straight line you can try to do a linear regression.

Press [STAT] Arrow right to highlight [CALC]

Select [4: LinReg(ax+b)]

LinReg(ax+b) echoes on run screen. Press [2nd] L1 Press comma, Press [2nd] L2 press comma Press [VARS] Arrow right to select [Y:VARs][1:Function] [ENTER][1:Y1]

ENTER

That is the equation for the linear regression. As you see eqaution has been saved in Y1. Press [GRAPH] to get the data and the regression curve drawn on same graph.

Hope it helps.

Oct 04, 2009 | Texas Instruments TI-84 Plus Calculator

I think you may find arcsin(x) is equivalent in older nomenclature to sin^-1 (x)...ie use the "2nd" and the SIN key instead of typing arcsin.

eg. arcsin(0.5) is 30 degrees is the same as sin^-1(0.5)

The ^-1 does not mean reciprocal, but "the angle whose sin is." Here the minus one indicates a kind of inverse operation. The word arcsin indicates that same inverse.

eg. arcsin(0.5) is 30 degrees is the same as sin^-1(0.5)

The ^-1 does not mean reciprocal, but "the angle whose sin is." Here the minus one indicates a kind of inverse operation. The word arcsin indicates that same inverse.

Mar 23, 2009 | Texas Instruments TI-89 Calculator

I had the same problem, this worked for me:

Mode -> page 2 -> Exact/Approx change from "Exact" -> "Auto"

Mode -> page 2 -> Exact/Approx change from "Exact" -> "Auto"

May 12, 2008 | Texas Instruments TI-89 Calculator

An argument must be in a specified domain. For example, ans(100) is not valid because the argument for ans() must be in the range 1–99.

May 11, 2008 | Texas Instruments TI-89 Calculator

1,393 people viewed this question

Usually answered in minutes!

exactly the same prob.

×