Fixing error (miscalculation) of the lags on page 130:

The start = start(ts.obj) + l argument add the number of lags to the cycle units of the start argument. For example:

> start(ts.obj) 
[1] 1976    1
> start(ts.obj) + l
[1] 1979    4

Solution - replacing start = start(ts.obj) + l with time(ts_merged)[l + 1]:

# The lags function return the series with its l lags
lags <- function(ts.obj, l){
  ts_merged <- NULL
  # Creating n lags
  for(i in 1:l){
    ts_merged <- ts.union(ts_merged, stats::lag(ts.obj, k = -i))
  }
  # Merge the lags with the original series
  ts_merged <- ts.union(ts.obj, ts_merged)
  # Set the columns names
  colnames(ts_merged) <- c("y", paste0("y_", 1:i))
  # Removing missing values as results of creating the lags
  ts_merged <- window(ts_merged,
                      start = time(ts_merged)[l + 1],
                      end = end(ts.obj))
  return(ts_merged)
}

Thanks for Anoushiravan Razmavar for pointing out this issue!

Hello. In Chapter 2, My RStudio cannot convert US_long_format and CA_mix_format.
Please advise. Thx!

My code and output is ..

str(dates_df)
'data.frame': 22 obs. of 7 variables:
$ Japanese_format : chr "2017/1/20" "2017/1/21" "2017/1/22" "2017/1/23" ...
$ US_format : chr "1/20/2017" "1/21/2017" "1/22/2017" "1/23/2017" ...
$ US_long_format : chr "Friday, January 20, 2017" "Saturday, January 21, 2017" "Sunday, January 22, 2017" "Monday, January 23, 2017" ...
$ CA_mix_format : chr "20-Jan-17" "21-Jan-17" "22-Jan-17" "23-Jan-17" ...
$ SA_mix_format : chr "20-Jan-17" "21-Jan-17" "22-Jan-17" "23-Jan-17" ...
$ NZ_format : chr "20/01/2017" "21/01/2017" "22/01/2017" "23/01/2017" ...
$ Excel_Numeric_Format: int 42755 42756 42757 42758 42759 42760 42761 42762 42763 42764 ...

dates_df$Japanese_format_new <-

• as.Date(dates_df$Japanese_format)

head(dates_df[ , c("Japanese_format", "Japanese_format_new")])
Japanese_format Japanese_format_new
1 2017/1/20 2017-01-20
2 2017/1/21 2017-01-21
3 2017/1/22 2017-01-22
4 2017/1/23 2017-01-23
5 2017/1/24 2017-01-24
6 2017/1/25 2017-01-25

identical(dates_df$Japanese_format, dates_df$Japanese_format_new)
[1] FALSE

class(dates_df$Japanese_format)
[1] "character"
class(dates_df$Japanese_format_new)
[1] "Date"

as.Date("31-01-2008")
[1] "0031-01-20"

as.Date("31-01-2008", format = "%d-%m-%Y")
[1] "2008-01-31"

dates_df$US_format[1]
[1] "1/20/2017"

dates_df$US_format_new <- as.Date(dates_df$US_format, format = "%m/%d/%Y")

dates_df$US_long_format[1]
[1] "Friday, January 20, 2017"
dates_df$US_long_format_new <- as.Date(dates_df$US_long_format,

•                                    format = "%A, %B %d, %Y")
  

dates_df$CA_mix_format[1]
[1] "20-Jan-17"
dates_df$CA_mix_format_new <- as.Date(dates_df$CA_mix_format,

•                                   format = "%d-%b-%y")
  

dates_df$NZ_format[1]
[1] "20/01/2017"
dates_df$NZ_format_new <- as.Date(dates_df$NZ_format,

•                               format = "%d/%m/%Y")
  

head(dates_df)
Japanese_format US_format US_long_format CA_mix_format
1 2017/1/20 1/20/2017 Friday, January 20, 2017 20-Jan-17
2 2017/1/21 1/21/2017 Saturday, January 21, 2017 21-Jan-17
3 2017/1/22 1/22/2017 Sunday, January 22, 2017 22-Jan-17
4 2017/1/23 1/23/2017 Monday, January 23, 2017 23-Jan-17
5 2017/1/24 1/24/2017 Tuesday, January 24, 2017 24-Jan-17
6 2017/1/25 1/25/2017 Wednesday, January 25, 2017 25-Jan-17
SA_mix_format NZ_format Excel_Numeric_Format Japanese_format_new
1 20-Jan-17 20/01/2017 42755 2017-01-20
2 21-Jan-17 21/01/2017 42756 2017-01-21
3 22-Jan-17 22/01/2017 42757 2017-01-22
4 23-Jan-17 23/01/2017 42758 2017-01-23
5 24-Jan-17 24/01/2017 42759 2017-01-24
6 25-Jan-17 25/01/2017 42760 2017-01-25
US_format_new US_long_format_new CA_mix_format_new NZ_format_new
1 2017-01-20 2017-01-20
2 2017-01-21 2017-01-21
3 2017-01-22 2017-01-22
4 2017-01-23 2017-01-23
5 2017-01-24 2017-01-24
6 2017-01-25 2017-01-25
Sys.info()
sysname release version nodename
"Windows" "10 x64" "build 22621" "LAPTOP-0CQKUS9G"
machine login user effective_user
"x86-64" "sato_" "sato_" "sato_"

my_err.txt

Hi Rami,

Onto the right repo.

Here's the block from the script:
My suspicion is that the problem is derived from hessian = TRUE.

data(USgas)

ts_plot(USgas,
title = "US Monthly Natural Gas consumption",
Ytitle = "Billion Cubic Feet",
Xtitle = "Year")

USgas_split <- ts_split(USgas, sample.out = 12)

train <- USgas_split$train
test <- USgas_split$test

par(mfrow=c(1,2))
acf(train, lag.max = 60)
pacf(train, lag.max = 60)

USgas_d12 <- diff(train, 12)

ts_plot(USgas_d12,
title = "US Monthly Natural Gas consumption - First Seasonal Difference",
Ytitle = "Billion Cubic Feet (First Difference)",
Xtitle = "Year")

USgas_d12_1 <- diff(diff(USgas_d12, 1))

ts_plot(USgas_d12_1,
title = "US Monthly Natural Gas consumption - First Seasonal and Non-Seasonal Differencing",
Ytitle = "Billion Cubic Feet (Difference)",
Xtitle = "Year")

par(mfrow=c(1,2))
acf(USgas_d12_1, lag.max = 60)
pacf(USgas_d12_1, lag.max = 60)

p <- q <- P <- Q <- 0:2

arima_grid <- expand.grid(p,q,P,Q)
names(arima_grid) <- c("p", "q", "P", "Q")
arima_grid$d <- 1
arima_grid$D <- 1

arima_grid$k <- rowSums(arima_grid)

library(dplyr)

arima_grid <- arima_grid %>% filter(k <= 7)

arima_search <- lapply(1:nrow(arima_grid), function(i){
md <- NULL
md <- arima(train, order = c(arima_grid$p[i], 1, arima_grid$q[i]), seasonal = list(order = c(arima_grid$P[i], 1, arima_grid$Q[i])))

results <- data.frame(p = arima_grid$p[i], d = 1, q = arima_grid$q[i],
P = arima_grid$P[i], D = 1, Q = arima_grid$Q[i],
AIC = md$aic)
}) %>% bind_rows() %>% arrange(AIC)

Error in optim(init[mask], armafn, method = optim.method, hessian = TRUE, :
non-finite finite-difference value [2]
Called from: optim(init[mask], armafn, method = optim.method, hessian = TRUE,
control = optim.control, trans = as.logical(transform.pars))