# Time series linear regression

Hello, I am fairly new to linearr regression and i want to see the relationship between two time series variables.I have two data variables: Mill rates and Increment values from 2014-2018 (i know it's not a lot of data). As increment value is created via TIF districts, the mill rate goes up. Incremental value is dollar amounts (i.e \$112,309,600) and mill rate is in decimals but represent percentages (i.e .024578). I know there are problems with autocorrelation when it comes to timeseries, but i used the tslm() which I hope solves that problem ?

Here is my code:
millratets <- ts(Millrate, start=2014)
incrementts <- ts(`Increment Value Generated From Previous year`, start=2014)
regts <- tslm(millratets~log(incrementts))
summary(regts)

(i chose level-log model because it would be easier to interpret)

Here are my results:
Call: tslm(formula = millratets ~ log(incrementts))

Residuals:
Time Series:
Start = 2014
End = 2018
Frequency = 1
1 2 3 4 5
-0.00018196 -0.00012986 0.00007215 0.00040960 -0.00016993

## Coefficients: ----------------------- Estimate --Std. Error-- t value-- Pr(>|t|) (Intercept) --------0.0250244 -- 0.0082064--3.049---0.05545 . log(incrementts) 0.0026857-- 0.0004363 -- 6.155 --0.00863 **

Signif. codes: 0 ‘’ 0.001 ‘’ 0.01 ‘’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.0002897 on 3 degrees of freedom
Multiple R-squared: 0.9266, Adjusted R-squared: 0.9022
F-statistic: 37.88 on 1 and 3 DF, p-value: 0.008629

At first glance, seems like good results to me and the coefficients make sense. For everyone 1% increase in incremental value the mill rate goes up .0026. I just wanted to double check with the geniuses on this forum and make sure everything was legit

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.