Find an Exponential Regression Using a Graphing Calculator

Revision as of 22:04, 12 May 2016 by Kipkis (Kipkis | contribs) (importing article from wikihow)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Sometimes, you will be given a series of points and want to find an exponential equation without having to resort to algebra. Luckily, you can use a graphing calculator to find the regression - it's easier than you'd think!

Steps

  1. Enter your points. On the TI83 and TI84, all editions, go to "STAT" then scroll down to "Edit" and press enter. You will be taken to the list editor. If there are already entries in L1 or L2, move your cursor over the text "L1" or "L2" and press clear, then press enter to delete all of the data. In L1, enter all of your x-coordinates in order pressing enter after each entry. In L2, enter all of your y-coordinates that correspond to the adjacent x-coordinates. Make sure they all match up so when you read from left to right, the ordered pairs are the same as your data.
  2. Go to the home screen. Press "2nd" then "MODE" to quit. Don't press clear or you might erase some of your data.
  3. Calculate the regression. Press "STAT" again then scroll right until you get to the "CALC" menu. Scroll down until you get to "ExpReg." then press enter twice. You should get the regression of the data you entered.
  4. Interpret the regression. At the top, you should see the equation the calculator is using. It will say y=a*b^x. The value of "a" is initial value (the y-intercept) and the value of "b" is 1 plus the rate. You can easily write out the equation and find out the rate and initial value from this data. Also be aware of the "r" value. The closer the value is to 1, the better fit the regression is to the data. If the value is less than or near .5, you may be better off with a different model. For anything significantly higher than .5, but less than .9, you might want to graph it and use your discretion as to whether or not the model fits.

Video

Related Articles