- This topic has 3 replies, 3 voices, and was last updated Apr-179:50 pm by
.
-
Posts
-
-
Up3
So i am revising quants, and i cant understand the difference between
i)Chebyshev’s inequality
ii)Confidence Interval
iii)Z-value
iv)t-valueThey all seem the same to me!
-
Up4
Yea. thanx.
this really helps.One thing though, how would we calculate z or t values? Will we hav tables? Or can we rely on not being tested on such calculations?
-
Up2
@aanchalb I am pretty sure they will provide the necessary tables if calculations are required. HOWEVER, you should know that the Z critical value for your confidence intervals: Z (1 stdev, or alpha=.1 AKA 90% confidence) ~ 1.645, Z (2 stdev, or alpha- .05 AKA 95% confidence) ~ 1.96, Z (3 stdev, or alpha=.01 AKA 99% confidence) ~ 2.58. These are easy test questions, and you should know them well.
-
Up0
Yeah, this happened to me earlier, confused the hell out of me. Not anymore though.
Okay first, Chebyshev’s inequality. It says that for any sample or population distribution, the % of observations that lie within k standard deviations of the mean is at leat 1 – 1/k^2 for every k greater than 1.
Second, confidence interval. It is a range of values that we expect the random variable to be in, for a certain percentage of time. It is also 1 – significance level and hence it can be stated as the probability of failing to reject a hypotheses when it is true or as 1 – Type 1 error.
Third, z-value (z-statistic) is used when the distribution is normal with a known variance whether the sample is small or large, i.e., lesser than 30 or greater than and equal to 30.
Fourth, t-value (t-statistic) is used when the distribution is normal with unknown variance for small and large samples and also when distribution is non-normal.
I hope this is what you were looking for.
-
- You must be logged in to reply to this topic.