Time series: data analysis and theory
Time series: data analysis and theory
Time Series Analysis, Forecasting and Control
Time Series Analysis, Forecasting and Control
| Hi-index | 0.00 |
The standard error of the sample mean for autocorrelated data is directly proportional to the value of the spectral density evaluated at zero frequency of the process being sampled. Thus confidence intervals for the true mean using traditional formulas can be greatly in error.This paper describes both parametric and nonparametric methods of spectral density estimation and illustrates numerically the basic results using a personal computer program called TIMESLAB which acts as a laboratory for studying such problems.The results of the paper indicate that in many situations, adjusting for auto-correlation is easily performed.