Topic: Time Series Finite Distributed Lag Model(FDL)
Use the fdl_model_data located in the Final Module of the Canvas to answer the following
question:
Suppose we are studying the impact of a certain policy (e.g., a marketing campaign) on sales
over time. We are interested in how the implementation of this policy in a given period affects
sales not only in the same period but also in subsequent periods. We will model this effect up to
the third lag (t-1, t-2, t-3).
The dataset will contain the following variables:
\begin{itemize}
\item Year: The year of observation.
\item Policy: A binary variable indicating whether the policy was implemented in that year (1 if
implemented, 0 otherwise).
\item Sales: The amount of sales in that year.
\end{itemize}
The Finite Distributed Lag model will be specified as follows:
Sales$_t = \beta_0 + \beta_1 Policy_t + \beta_2 Policy_{t-1} + \beta_3 Policy_{t-2} + \beta_4 Policy_{t-3} + u_t$
where:
\begin{itemize}
\item Sales$_t$ is the sales in year t.
\item Policy$_t$, Policy$_{t-1}$, Policy$_{t-2}$, and Policy$_{t-3}$ are the current and lagged values of the policy
variable.
\item $\beta_0$, $\beta_1$, $\beta_2$, $\beta_3$, and $\beta_4$ are the coefficients to be estimated.
\item $u_t$ is the error term.
\end{itemize}
This dataset includes the original Policy variable and its lags up to three years (Policy_Lag1,
Policy_Lag2, Policy_Lag3). The Sales column represents the sales for each year.
A. Perform the FDL regression analysis on this dataset and interpret the results
B. Provide your Python Codes.
