2-14 Measurement error variance. We have a measuring device for which we are to ascertain that its measurement error variance is less than the borderline acceptable value of 100, You

can assume that the errors are zero-mean. We carry out N independent trials and the result is that the estimate of the variance is ở^2 = 80. For what N are you willing to risk your job by saying that there is a 5% (or less)probability to get such an estimate while the true value is o^2 = 100 (or higher)? Hint: You can assume a Gaussian distribution for the error in the estimate of the variance and use a probability region about the borderline value.