# Thread: Help with Regression Analysis problem and time delay

1. ## Help with Regression Analysis problem and time delay

Good morning all!

I am presently working on a paper, and am grappling with a statistical problem that is outside of my area of expertise (I am using Excel).

My dependent variables are recycling rate and program cost, and my independent variable is municipal funding.

I want to test how municipal funding in a previous time period effects recycling rate and program cost in the current time period

Example:

2011 City A Recycling Rate: 80%
2011 Program Costs: \$1,000,000
2010 Municipal Funding: \$40,000

I have this data for 223 cities over a 10 year time period, and I want to see how municipal funding over this period has effected recycling rate performance and cost containment.

Many thanks for the help!

2. ## Re: Help with Regression Analysis problem and time delay

Just specify a lag function: y_t= a+〖βx〗_(t-1)

My advise; don't use Excel, and read some sources about lag-functions. The answer to your question is quite simple.

3. ## Re: Help with Regression Analysis problem and time delay

Hi
First of all sorry, because the following is little bit off topic:

When statisticians talks to mathematicians about modeling, often the term lag is used but there is sometimes a confusion what “lag” actually means. From the pure meaning of the word “lag”, it is similar to a delay or retardation. Unfortunately, in math extreme caution has to be carried out because in some areas a “lag” is not a “delay”.

So therefore, could you please explain what statisticians exactly mean when they talk about a “lag function”. Is there a definition? That would be very helpful for me.

Thx!

4. ## Re: Help with Regression Analysis problem and time delay

See Doyle & Saunders (1985): The Lead Effect of Marketing Decisions. This source has in my opinion a good definition of lag/lead effects (if you have access to journals).

What you are trying to do is finding the effect of previous municipal spending (so, at time-1) on current values of recycling rate (so, at t=0). Hence, make two columns ,preferable in a statistical program, in which the first column has the recycling rate per year, and the second column the municipal spending from previous year (so shift them upwards). Thus, on one row should be the actual values of recycling rate and the value of municipal spending from previous year. Now you can use a regression analysis, where you use the first column as DV and the second column as IV.
Hope this is a bit clear.

*edit; ah see that the 2nd question was from another person.. nevertheless, hope above gives answer to both questions..

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts