Question 296799
Assuming your data is representative and randomly selected, then since the sample is large (>30) and estimating averages, the CLT allows you to use normal approximation to estimate the population mean

confidence intervals for the population mean via normal approximations are: point estimate +/- Noise
where the point estimate is the sample average and Noise= Z*Standard Error
or basically  Confidence interval for mean :  Xbar +/-Z*S/Sqrt(n)

The sample of 144 stores indicates that Xbar = 375 and S=20
The Z value corresponding to 99% is the number of standard deviations which capture 99% of the normal distribution in this case +/-2.576

So.  99% confidence interval for the population mean = 375-2.576*20/Sqrt(144) and 375+2.576*20/Sqrt(144)

The 95% confidence interval for the population mean is done the same way but substituting the Z =1.96