Hello all,
I am currently researching the impact of the equity valuation by the Shiller P/E ratio on different equity strategies on an equity index.
I made a scatter plot where I plotted the ten year forward annualized returns with the corresponding equity valuation at that moment. For the period 1995-2005 i have the ten year annualized returns for each month, so 120 data points per strategy. I plotted trend lines for each strategy in order to try to forecast the ten year forward annualized return per strategy with the current level of equity valuation.
I let excel automatically show the formula of the trend lines together with the r-squared (R2). Now I know that r-squared is not the most reliable assessment of significance and I want to 'test' R2. I would also like to use a different measurement (standard error?, p-value? confidence interval?) to test the statistical significance of the plotted trend lines, in order to know how reliable my predictions of future returns are.
With the data analysis tool 'regression' I get all sorts of information but I don't really understand what information to use and how to interpret these numbers.
If any one could help me I would be very thankful.
First image is the scatter plot, second image is the output of the data analysis tool Regression of excel.
I am currently researching the impact of the equity valuation by the Shiller P/E ratio on different equity strategies on an equity index.
I made a scatter plot where I plotted the ten year forward annualized returns with the corresponding equity valuation at that moment. For the period 1995-2005 i have the ten year annualized returns for each month, so 120 data points per strategy. I plotted trend lines for each strategy in order to try to forecast the ten year forward annualized return per strategy with the current level of equity valuation.
I let excel automatically show the formula of the trend lines together with the r-squared (R2). Now I know that r-squared is not the most reliable assessment of significance and I want to 'test' R2. I would also like to use a different measurement (standard error?, p-value? confidence interval?) to test the statistical significance of the plotted trend lines, in order to know how reliable my predictions of future returns are.
With the data analysis tool 'regression' I get all sorts of information but I don't really understand what information to use and how to interpret these numbers.
If any one could help me I would be very thankful.
First image is the scatter plot, second image is the output of the data analysis tool Regression of excel.
Last edited:
