# Thread: Analyzing 311 Calls trend

1. ## Analyzing 311 Calls trend

For an assignment, I am analyzing 311 call trends. I have downloaded the total number of 311 calls made in my city per month for the last five years and trying to project the remaining number of 311 calls I expect to be made in 2014. 311 calls tend to peak, at least from what I've observed, from Oct-Mar, and then fall off. The activity also tends to increase year-to-year. I am working on this in Excel and can show the trends in line graphs but what statistical functions in Excel can I use to quantify the trend? And then how do I actually make a projection? It doesn't have to be anything too precise. I just need something.

2. ## Re: Analyzing 311 Calls trend

This sounds like a time series analysis with a trend and seasonality. I don't think Excel has any built-in functionality to support this beyond trending and regression, which cannot handle the seasonality.

3. ## Re: Analyzing 311 Calls trend

Originally Posted by Miner
This sounds like a time series analysis with a trend and seasonality. I don't think Excel has any built-in functionality to support this beyond trending and regression, which cannot handle the seasonality.
So how would you do that in Excel? I see regression under the Data Analysis add-in. But what would the "Input Y-Range" and "Input X-Range" be?

4. ## Re: Analyzing 311 Calls trend

The best answer is not to do it in excel. You can create dummy variables for season (11 for months 3 for quarterly data) and a trend variable and run linear regression but if you have autocorrelation (which is very likely) your p values will probably be wrong.

Time series is simply too complex to be handled by excel which does basic stats.

It is possible to deseasonalize and detrend data an then analze that. I am unsure if you can do this in excel.

 Tweet

#### Posting Permissions

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