Lag functions

Module for producing lagged versions of pandas DataFrames

Let’s start out by defining a short multivariate time-series:

X = pd.DataFrame([[2,3,4],[5,6,7],[8,8,6],[9,10,3],[11,4,6]]); X

	0	1	2
0	2	3	4
1	5	6	7
2	8	8	6
3	9	10	3
4	11	4	6

---

source

lag

 lag (X:pandas.core.frame.DataFrame, lags:list)

Return a dataframe with specific lags

	Type	Details
X	DataFrame	A pandas dataframe with observations as rows and variables as columns
lags	list	A list with the same number of lists as the number of columns of X containing the desired lagged versions desired for the variable in that column
Returns	DataFrame	A dataframe with the specified lagged variables of X

lags = [[0,1],[0,2],[0,3]]

lag(X,lags)

	0		1		2
	t	t-1	t	t-2	t	t-3
3	9	8.0	10	6.0	3	4.0
4	11	9.0	4	8.0	6	7.0

---

source

lag_uniform

 lag_uniform (X:pandas.core.frame.DataFrame, lag:int)

	Type	Details
X	DataFrame	A pandas dataframe with observations as rows and variables as columns
lag	int	Integer of the maximum lag to include for all variables of X
Returns	DataFrame	A dataframe with the lagged variables of X

lag_uniform(X,2)

Variable	0	1	2	0	1	2	0	1	2
Time	t	t	t	t-1	t-1	t-1	t-2	t-2	t-2
0	8	8	6	5	6	7	2	3	4
1	9	10	3	8	8	6	5	6	7
2	11	4	6	9	10	3	8	8	6