btsdemo: Randomly Generated Panel Dataset

In mbannert/panelaggregation: Aggregate Longitudinal Survey Data

Description

This data was created by simulation to mimmick a firm level dataset stemming from business tendency surveys. The data was simulated because of privacy concerns with micro level firm data. For convenience the dataset contains two different date notations. Also 5 qualitative 3-item questions are included. Business tendency survey data is often weighted with company size represented by the number of employees. Thus the weight column is quantitative and its distribution is somewhat (!) reasonable with respect to the distribution of employees in a typical firm sample.

Format

A data frame with 27000 rows and 13 variables

Details

• uid unique company identifier

• year numeric year column

• weight quantitative weight

• question\_1

• question\_2

• question\_3

• question\_4

• question\_5

• group group to mimmick different sectors / branches of trade

• altGroup another alternative grouping columns

• sClass a column denoting discrete size classes small (S), medium (M) and large (L)

• date\_qtrly quarterly dates stored in a single column.

Author(s)

Matthias Bannert

Source

Randomly generated in R using the sample generator from https://github.com/mbannert/gateveys/blob/master/R/gateveys.R

mbannert/panelaggregation documentation built on May 21, 2019, 2:24 p.m.