Using the commuter airline data, fit the no change model and two other naive mod
ID: 3271753 • Letter: U
Question
Using the commuter airline data, fit the no change model and two other naive models. Evaluate the fit of these three models by various statistics inclusive of MSE, RMSE, MAD,and MAPE. Comment on which model is best and why. Don't forget to graph the data and forecast results for your best choice.
Trend Commute Majair Year
1 12.83 2.91 1982
2 6.87 4.2 1983
3 8.22 2.29 1984
4 7.03 2.8 1985
5 5 2.89 1986
6 11.74 4.25 1987
7 6.19 3.54 1988
8 6.74 2.75 1989
9 4.75 2.44 1990
10 8.15 2.8 1991
11 7.06 2 1992
12 4.44 2.85 1993
13 2.79 2.17 1994
14 3.73 3.7 1995
15 3.13 3.95 1996
16 11.48 4.33 1997
17 11.31 3.89 1998
18 19.34 3.68 1999
19 19.88 4.43 2000
20 12.54 3.48 2001
21 13.63 3.31 2002
22 3.49 4.99 2003
23 7.43 2.13 2004
24 11.38 3.12 2005
25 5.28 2.45 2006
26 5.06 2.42 2007
27 11.89 1.95 2008
28 3.39 2.72 2009
29 9.91 2.96 2010
30 6.58 3.08 2011
31 8.31 2.81 2012
32 12.18 2.07 2013
33 6.35 3.14 2014
34 8.29 3.05 2015
Explanation / Answer
We cas analysis this by R.
library(ggplot2)
library(ggfortify)
library(forecast)
x=read.csv("Book1.csv")
tseries=ts(data = x, start = 1982, end = 2015, frequency = 1)
noc=lag(tseries, k = 1) ## no change
a=accuracy(noc, tseries)
p=0.05
pc=(1+p)*(noc) ## percentage change
b=accuracy(pc, tseries)
l2=lag(tseries, 2)
prc=noc + (1 + p)*(noc-l2) # proportion change
c=accuracy(prc, tseries)
y=data.frame(a,b,c)
y
ME RMSE MAE MPE MAPE ACF1 Theil.s.U
Noc -0.4666 2.5104 1.62439 -8.9886 22.94859 -0.3667306 0.007566704
ME.1 RMSE.1 MAE.1 MPE.1 MAPE.1 ACF1.1 Theil.s.U.1
pc -25.821 50.54561 26.87633 -14.43808 26.66836 0.962452 0.05122932
ME.2 RMSE.2 MAE.2 MPE.2 MAPE.2 ACF1.2 Theil.s.U.2
prc 0.07604297 4.194903 2.118559 -5.460753 36.34252 -0.6828871 0.009229979
Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.