1 (Mx1 by N) to (MxQ by N+1)

Go to the RMD, R, PDF, or HTML version of this file. Go back to fan’s REconTools Package, R Code Examples Repository (bookdown site), or Intro Stats with R Repository (bookdown site).

Case One: There is a dataframe with \(M\) rows, based on these \(m\) specific information, generate dataframes for each \(m\). Stack these indivdiual dataframes together and merge original \(m\) specific information in as well. The number of rows for each \(m\) is \(Q_m\), each \(m\) could have different number of expansion rows.

Generate a panel with \(M\) individuals, each individual is observed for different spans of times (uncount). Before expanding, generate individual specific normal distribution standard deviation. All individuals share the same mean, but have increasing standard deviations.

1.1 Generate Dataframe with M Rows.

This is the first step, generate \(M\) rows of data, to be expanded. Each row contains the number of normal draws to make and the mean and the standard deviation for normal daraws that are \(m\) specific.

# Parameter Setups
it_M <- 3
it_Q_max <- 5
fl_rnorm_mu <- 1000
ar_rnorm_sd <- seq(0.01, 200, length.out=it_M)
ar_it_q <- sample.int(it_Q_max, it_M, replace=TRUE)

# N by Q varying parameters
mt_data = cbind(ar_it_q, ar_rnorm_sd)
tb_M <- as_tibble(mt_data) %>% rowid_to_column(var = "ID") %>%
                rename(sd = ar_rnorm_sd, Q = ar_it_q) %>%
                mutate(mean = fl_rnorm_mu)

# display
kable(tb_M) %>%
  kable_styling_fc()
ID Q sd mean
1 1 0.010 1000
2 3 100.005 1000
3 5 200.000 1000

1.2 Random Normal Draw Expansion

The steps are:

  1. do anything
  2. use “.$” sign to refer to variable names, or [[‘name’]]
  3. unnest
  4. left_join expanded and original

Note these all give the same results

Use dot dollar to get variables

# Generate $Q_m$ individual specific incomes, expanded different number of times for each m
tb_income <- tb_M %>% group_by(ID) %>%
  do(income = rnorm(.$Q, mean=.$mean, sd=.$sd)) %>%
  unnest(c(income))

# Merge back with tb_M
tb_income_full_dd <- tb_income %>%
  left_join(tb_M)

# display
kable(tb_income) %>%
  kable_styling_fc()
ID income
1 999.9927
2 978.4124
2 966.5070
2 891.4247
3 982.9153
3 1214.1221
3 970.9213
3 766.8910
3 836.2969 
kable(tb_income_full_dd) %>%
  kable_styling_fc()
ID income Q sd mean
1 999.9927 1 0.010 1000
2 978.4124 3 100.005 1000
2 966.5070 3 100.005 1000
2 891.4247 3 100.005 1000
3 982.9153 5 200.000 1000
3 1214.1221 5 200.000 1000
3 970.9213 5 200.000 1000
3 766.8910 5 200.000 1000
3 836.2969 5 200.000 1000