@Susandoyle @RealSeanBrodrick it's just a scatter plot so there is no statistical test. We do need to know where and how the prices are drawn because we're discretizing a continuous function. If it's something like EOD average price of gas in the US vs price for a barrel of crude, OK. I suspect there's not enough intraday swing to have an effect. I would like to see the addition of other variables, like refining and supply chain capacity to help explain any correlation (what is the r btw)?
@Susandoyle @RealSeanBrodrick @Susandoyle @RealSeanBrodrick actually now that I think about it, as Susan intimated, this plot is almost meaningless. To see what I mean, without further info, one can conclude that *every* time in the past 2 years say, oil was at $105, gasoline was at $3.75.
@hallmarc also are we seeing only a specific part of the scatter plot, ie a 25% quartile. @RealSeanBrodrick
@hallmarc @RealSeanBrodrick was it strictly a regression analysis? I need to know these things
@RealSeanBrodrick I would need to see the statistics behind this scatter plot before discussing. What was the sample made of? Sample size? What statistical test was used? What are the p value? Effect? Confidence intervals?