**New here?**Learn about Bountify and follow @bountify to get notified of new bounties! Follow @bountify x

I want to make this example work with time series data:

https://bl.ocks.org/ctufts/298bfe4b11989960eeeecc9394e9f118

My x-values are epoch dates that are one day apart from the next one (representing each day of the past month, for example). My y-values are integers from 0 to 10.

Please modify the example to produce sample data within that range, and to draw a meaningful regression line.

# Edit

Here is some sample data. Included are the yhat values if you simply use those x and y values. Notice how the yhat values (which what get plotted on the graph) are all very similar, which results in a near-horizontal regression line, no matter the values. That's the problem.

The reason the yhat values are similar is probably because the x-values here are all so close to the mean value. Look at the original example at bl.ocks.org: x values are from 0 to 1000. In this data set, x values are from 1480536255 to 1481313855.

So I think you'd have to change how the yhat values are calculated, or normalize the x-values before calculating them.

```
[{:x=>1480536255, :y=>7, :yhat=>7.098095616083853}, {:x=>1480622655, :y=>8, :yhat=>7.098509840067301}, {:x=>1480709055, :y=>8, :yhat=>7.098924064050747}, {:x=>1480795455, :y=>6, :yhat=>7.099338288034195}, {:x=>1480881855, :y=>8, :yhat=>7.099752512017643}, {:x=>1480968255, :y=>6, :yhat=>7.100166736001091}, {:x=>1481054655, :y=>6, :yhat=>7.100580959984537}, {:x=>1481141055, :y=>8, :yhat=>7.100995183967985}, {:x=>1481227455, :y=>7, :yhat=>7.101409407951433}, {:x=>1481313855, :y=>7, :yhat=>7.101823631934881}]
```

## 1 Solution

Hello

I've made a test with your values based on the d3js code you proposed, the only difference is that the values are not randomly generated, but your ones, and recentered the x values arount the "zero" (line 139 to 154). Just copy/paste the following code into an HTML file and it should do the job.