[R] Puzzled at lm() and time-series

[Gabor Grothendieck]

Ajay Shah <ajayshah <at> mayin.org> writes:
: 
: But I can't get an OLS going:
: 
: > lm(g.cpi.iw ~ g.wpi, M)
: Error in "storage.mode<-"(`*tmp*`, value = "double") : 
:         invalid time series parameters specified
: 

: 
: M = structure(list(g.wpi = structure(c(11.14, 11.83, 11.88, 12.66,
: 13.2, 14.54, 16.18, 16.88, 16.88, 11.02, 10.94, 9.7, 9.61, 8.94, 8.92,
: 8.47, 8.24, 6.62, 5.04, 4.51, 4.53, 3.67, 3.58, 3.61, 4.28, 4.9, 5.08,
: 4.54, 4.48, 5.25, 5.12, 5.47, 5.43, 5.86, 5.09, 5.07, 3.6, 3.26, 3.83,
: 4.38, 3.94, 4.01, 5.11, 4.17, 4.32, 4.57, 5.64, 6.43, 7.1, 6.56, 5.89,
: 6.5, 7.15, 6.32, 4.48, 5.4, 5.41, 3.98, 3.31, 2.49, 1.97, 2.79, 3.23,
: 3.41, 3.1, 2.79, 3.58, 3.5, 5.54, 6.56, 6.35, 6.5, 6.51, 6.12, 6.47,
: 7.54, 7.61, 8.31, 8.69, 8.35, 6.45, 5.42, 5.6, 5.35, 5.27, 5.38, 4.48,
: 2.89, 2.59, 2.26, 1.53, 1.35, 1.72, 1.5, 1.5, 2.4, 2.79, 3.35, 3.55,
: 3.08, 3.39, 3.31, 4.21, 5.38, 6, 6.62, 6.57, 5.39, 4.68, 3.96, 4.9,
: 5.15, 5.42, 5.79, 6.45, 6.14, 4.79, 4.52, 4.99, 6.14, 6.86), .Tsp =
: c(1994.5, 2004.5, 12), class = "ts"), g.cpi.iw = structure(c(11.07,
: 10.94, 11.2, 10.31, 9.81, 9.47, 9.89, 9.81, 9.74, 9.67, 10.29, 10.47,
: 11.39, 10.92, 10.07, 10.38, 10.31, 9.69, 9, 8.59, 8.87, 9.83, 9.33,
: 8.82, 8.31, 8.89, 8.52, 8.46, 8.72, 10.41, 11.11, 10.76, 10.03, 9.26,
: 7.32, 6.61, 5.6, 4.66, 4.94, 5.49, 4.87, 6.29, 9.71, 9.14, 8.26, 8.19,
: 10.51, 12.39, 14.8, 15.04, 16.34, 18.63, 19.67, 15.32, 9.38, 8.64,
: 8.95, 8.36, 7.71, 5.26, 3.16, 3.15, 2.14, 0.92, 0, 0.47, 2.62, 3.61,
: 4.83, 5.54, 5.01, 5.24, 4.95, 3.99, 3.5, 2.75, 2.74, 3.48, 3.25, 3.02,
: 2.53, 2.28, 2.5, 3.39, 4.04, 5.19, 4.73, 4.23, 4.89, 5.16, 4.94, 5.19,
: 5.17, 4.69, 4.66, 4.16, 3.89, 3.86, 4.3, 4.06, 3.6, 3.2, 3.43, 3.86,
: 4.06, 5.12, 4.66, 4.41, 4.16, 3.1, 2.89, 3.29, 3.07, 3.72, 4.35, 4.13,
: 3.49, 2.23, 2.83, 3.02, NA), .Tsp = c(1994.5, 2004.5, 12), class =
: "ts")), .Names = c("g.wpi", "g.cpi.iw"), row.names = c("1", "2", "3",
: "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15",
: "16", "17", "18", "19", "20", "21", "22", "23", "24", "25", "26",
: "27", "28", "29", "30", "31", "32", "33", "34", "35", "36", "37",
: "38", "39", "40", "41", "42", "43", "44", "45", "46", "47", "48",
: "49", "50", "51", "52", "53", "54", "55", "56", "57", "58", "59",
: "60", "61", "62", "63", "64", "65", "66", "67", "68", "69", "70",
: "71", "72", "73", "74", "75", "76", "77", "78", "79", "80", "81",
: "82", "83", "84", "85", "86", "87", "88", "89", "90", "91", "92",
: "93", "94", "95", "96", "97", "98", "99", "100", "101", "102", "103",
: "104", "105", "106", "107", "108", "109", "110", "111", "112", "113",
: "114", "115", "116", "117", "118", "119", "120", "121"), class =
: "data.frame")
: 

>From the above we see there is an NA in the last observation.
If we delete that observation then it runs to completion:

R> lm(g.cpi.iw ~ g.wpi, M, subset = -121)

Call:
lm(formula = g.cpi.iw ~ g.wpi, data = M, subset = -121)

Coefficients:
(Intercept)        g.wpi  
     3.7952       0.5083