My name is Vlad Obukhanich, and I am a rising junior at Boston University who is really interested in quantitative finance. I am a part of the quant research team within the BU Finance and Investment Club, where we use the Quantopian Research Environment for educational purposes.
Recently, I decided to look at how the US unemployment rate affects stock market performance. In particular, I wanted to test the claim that historically SPY returns were the highest when the US unemployment rate was 9% or above (http://awealthofcommonsense.com/2014/06/unemployment-rate-stock-performance/). I created a notebook to conduct the following hypothesis test: 'Are the SPY returns when the unemployment rate was 9% or above significantly higher than the SPY returns when the unemployment rate was below 9%?'
Just wanted to share the results of my research with the community and possibly receive some feedback on what are some things that could be improved about it. I also have not come up with any implications of my research yet as I am not sure if I can make any sound conclusions that are backed by only 15 years worth of data (the earliest year that was available to me in the US Civilian Unemployment Rate dataset was 2002).
Also, if you are curious about other hypothesis tests similar to mine, feel free to check out the following posts: