自相关：

If we denote the error term as u_i对于我^thobservation, then we assumed that:

其中cov（x，y）是两个变量x和y'之间的协方差。但是，如果违反了此假设，则存在自相关。那是，

There are typically two types of data used for empirical studies: cross section and time series. In the first case data on various variables are collected across a number of households or firms at a given point in time. Examples in India include the Economic Census, and the Census of India. This type of data can be used to test for any relationship between variables across firms or households (depending on the sampling unit). For instance, are bigger firms (in terms of revenue) more productive (Per unit of labour) is a question that can be answered using the' Economic Census data. Data may also be collected over time for a set of variables for the same sampling unit (usually at country or state level, may also be for households or firms). This type of data is known as time series data. This allows us to check for any relationship between the variables for the same unit, over time. For instance, if aggregate consumption and income data are collected at the country level, then we can estimate the consumption function, that is, the relationship between income and consumption. As you know from macroeconomic analysis, here we can answer questions such as: Does consumption increases with income and if so by what amount? Another form of data that has gained prominence in recent times is when time series data is available for multiple units, then we say this is pooled (time series and cross-section data), also called panel data.