Sir, why are there 2 different techniques for hypothesis testing when you said that either of them will give us the same answer? Is it because the interval estimation approach is more accurate due to low level of type 2 error?
@muskankaur85403 жыл бұрын
Sir, can you give more example on explaining the p value.
@SinghSajal Жыл бұрын
Can someone advise from where the value of Tabulate t came from ?? 2.306 ?? Please I have an exam, can someone throw some light ?
@ritasreede7460 Жыл бұрын
You get a t table, where you can find out the value according to the probability and the degrees of freedom. Since this is at 5% LOS, you look at the column with probability 0.975 (1-0.025)
@SnehalBhartiyaАй бұрын
abe google search kar t-table
@michaeljosvawchristensen18562 жыл бұрын
Nice material❤ why is t statistic = beta hat divided by se ? I thought it should be beta hat minus H0 condition value and then divided by we ?
@Priyanka-ll8pq Жыл бұрын
Under H0, beta = 0 So, t statistic = (beta hat - beta)/s.e.(beta hat) leads to (beta hat - 0)/s.e.(beta hat) under the condition of null hypothesis.
@shivakumaram58022 жыл бұрын
p*100>10 leads to insignificant(accept null) , means p value is more which tells that probablity of occurance of type 1 error is more ,then how we can conclude that there ll be insignificant.(objective is to minimise type 1 error)?????
@Priyanka-ll8pq Жыл бұрын
Our objective is to minimize the type 1 error (or likely preferred to be kept at a pre-specified level, say, 1%, 5% or 10%), while rejecting null hypothesis. Since for level of significance >10 implies that while rejecting the null hypothesis, we will probably commit >10 type 1 errors per 100 samples, we accept the null hypothesis, which will likely be the correct decision, i.e., not rejecting H0 when H0 is true. I would refer you to the previous video of this series to understand it better. The professor showed a table to easily explain Type-1 and Type-2 error.