I did notice that the three coefficients are very close but not quite identical to those derived by using the graphical trendline feature under the charts tab. Lastly, in the next cell over, again type the same formula but change the last number to a 3. In an adjacent cell, type the same formula again but change the final 1 to a 2. of course you'd change these ranges appropriately to match your data). Using the formula: =LINEST(B2:B21, A2:A21^1.2)Įxcel's LINEST function can also calculate multiple regressions, with different exponents on x at the same time, e.g.: =LINEST(B2:B21,A2:A21^,TRUE,FALSE),1) (by the way, the B2:B21 and A2:A21 I used are just the same values the first poster who answered this used. You can also use Excel to calculate a regression with a formula that uses an exponent for x different from 1, e.g. Which returns a single value, the linear slope ( m) according to the formula: Check the spelling of the file name, and verify that the file locations is correct.
#EXCEL LINEAR REGRESSION ERROR HOW TO#
i would like to ask some questions on how to derive some of the quantities listed in the last table. You need to use an undocumented trick with Excel's LINEST function: =LINEST(known_y's,, , )Ī regular linear regression is calculated (with your data) as: =LINEST(B2:B21,A2:A21) In regression analysis, Excel calculates for each point the squared difference between the y-value estimated for that point and its actual y-value.