From: tjroberts137@sbcglobal.net   
      
   On 3/27/19 6:10 AM, Savin Beniwal wrote:   
   > Let's say we performed an experiment using some theoretical model and   
   > observational data points. Our aim is to get the best fit values of   
   > unknown parameters with their error using observations. Using some   
   > statistical techniques i.e. least square fitting or chi-square or MCMC,   
   > we got the best fit values and error for each parameter. Suppose the   
   > error of parameters is higher than their best-fit values   
   > i.e. [A=0.2+-1.5] here the best-fit value is 0.2 and error is 1.5.   
   > In this scenario what could you conclude from this experiment? Please   
   > put your comments and thoughts.   
      
   You can conclude that the result is consistent with A being zero. If the   
   theoretical value of A is anywhere -2 < A < 2 or so, you can conclude   
   that the experiment is consistent with the prediction of the theory.   
      
   A famous example of this is the Michelson-Morley experiment of 1887.   
   Note that in 1887 statistical error analysis was not known, and   
   errorbars were not used. One can look at their fig. 6 and imagine there   
   is a variation consistent with the predicted sinusoid, and there is a   
   large literature of non-physicist authors claiming M&M "saw the ether"   
   [#]. But each point in that plot is an average of 20 values, from which   
   one can estimate errorbars, and they are roughly five times larger than   
   the variations. So a modern conclusion is that they did not see any   
   significant variation of fringe position with orientation, fully   
   consistent with the prediction of SR.   
      
   	[#] There are many more papers about the similar but much   
   	more voluminous measurements of D. C. Miller in the 1920s-   
   	30s. A modern error analysis shows his measurements also   
   	display no significant variation with orientation, and   
   	are fully consistent with the prediction of SR. Unlike   
   	M&M, Miller's raw data have been preserved.   
   	https://arxiv.org/abs/physics/0608238   
      
   Tom Roberts   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)