Misc. Stats Topics Discussion > Relationship between 3-sigma rule and t-statistic
Laura,
A t-statistic is standardized indicator of whether a given parameter is significantly different from zero. A common use for a t-statistic is to test whether differences between means are greater than zero or whether a coefficient (in a regression) is significantly larger than zero. The t-statistic is roughly derived by the ratio of a parameter to its standard error.
The "3-sigma rule" refers to the principle that most values in a normal distribution are within 3 standard deviations (or sigmas) from the mean (approximately 99.7% are within 3 standard errors, in a normal distribution). I hope this is helpful!
August 16, 2011 |
Jeremy Taylor
Jeremy Taylor
Is there a direct relationship between the 3-sigma rile and a t-statistic? If so, can someone explian this relationship? Cheers