Conference PaperPDF Available

Are IPCC climate-forecasting methods too complex?

  • June 2015
  • Conference: The 35th International Symposium on Forecasting (ISF 2015)
  • At: Riverside, USA
  • Volume: Proceedings
Authors:
Kesten Green at University of South Australia
Kesten Green
J. Scott Armstrong at University of Pennsylvania
J. Scott Armstrong
Download full-text PDF
Read full-text
Download citation

Abstract and Figures

The United Nations Intergovernmental Panel on Climate Change’s (IPCC’s) forecast of dangerous manmade global warming is an important influence on public policy decisions in many countries. Rational policies require that policy makers obtain forecasts from evidence-based methods. A review of comparative studies found that sophisticatedly simple methods, which can be readily understood by decision makers, provide forecasts that are substantially more accurate than those from more complex methods. Using a checklist of nine items, we assessed the simplicity of the IPCC’s forecasting procedures. The checklist items cover the understandability of the forecasting process and models, and the clarity of the relationships between the model outputs, forecasts, and decisions. Our checklist responses gave the IPCC procedures a “simplicity rating” of 1.7 or 19%. In contrast, a “no-trend forecast,” obtained a simplicity rating of 8.4 or 93%. We suggest that making policy decisions on the basis of forecasts from poorly understood procedures could be characterized as “faith-based policy making.”
No caption available
… 
No caption available
… 
Figures - uploaded by Kesten Green
Author content
All figure content in this area was uploaded by Kesten Green
Content may be subject to copyright.
Content uploaded by Kesten Green
Author content
All content in this area was uploaded by Kesten Green on Nov 27, 2015
Content may be subject to copyright.
!"#$%&"#'()*+,$-#./&0)$.&&$
'&-12#34
5#).#+$67$8"##+!"#University*of*South*Australia
97$:'&;$!"-)."&+,!"#University*of*Pennsylvania#
!Ehrenberg7Bass*Ins:tute*at*University*of*South*Australia
$%&'(%)*+%),#-./0+123/#+%#4+('5)1*%6
728'(129'"#:),2;+(%2)
<=#>3%'#<?@A
Slides*available*at*ForPrin.com
G&A$ISF$2015$–$Complex4V13
1
Thursday, 23 July 15
<+$./#$=(2>#$&%$'&-12#3$%&"#'()*+,$-#./&0)
B%),.1&1#C)8'#,+%6#)113/'9#&C)&#5+/0,'D#/'&C+91#)('#
%''9'9#&+#9'),#E2&C#5+/0,'D#0(+F,'/1G#
B1#,+%6#)6+#)1#@HIA#JB(/1&(+%6"#00G<<AK<L<M"#)#('82'E#+;#
&C'#,2&'()&3('#;+3%9#&C)&##5+/0,'D2&.#&'%91#&+N
•reduce#)553()5.#)%9#3%9'(1&)%92%6"#)%9
•increase#5+1&1#)%9#/21&)O'1
P'102&'#&C'#'829'%5'"#;+('5)1&'(1#5+%*%3'#&+#31'#'8'(#
/+('#5+/0,'D#/'&C+91G#
Q'#('82'E'9#&C21#2113'#F.#+(6)%2R2%6#)#JBR#-0'52),#$113'#+%#
Simple*versus*complex*forecas:ngG#Q'#('0+(&#+%#+3(#0)0'(#
2%#&C)&#2113'#('82'E2%6#&C'#'829'%5'G#
<
2
Thursday, 23 July 15
L
See$
HANDOUT
3
Thursday, 23 July 15
?#=@#A$&%$#31#"@-#+.(2$"#)#("'/
7'82'E'9#03F,21C'9#('1')(5C#;(+/#),,#)(')1#
+;#;+('5)1*%6"#2%5,392%6#&C'#-0'52),#$113'#
0)0'(1S#
@G9'T%'9#12/0,252&.#2%#;+('5)1*%6
<G29'%*T'9#1&392'1#E2&C#'829'%5'#+%#
5+/0)()*8'#)553()5.
LG)11'11'9#92('5*+%),#)%9#'U'5&#12R'#
'829'%5'
=
4
Thursday, 23 July 15
:@-12#$%&"#'()*+,$0#B+#0
-2/0,252&.#2%#;+('5)1*%6#('V32('1#&C)&#)('#all*of*the*
following*are#understood*by*clientsG#
@G#;+('5)1*%6#/'&C+9"
<G#('0('1'%&)*+%#+;#53/3,)*8'#O%+E,'96'"#
LG#(',)*+%1C201#2%#/+9',1"
=G#(',)*+%1C201#)/+%6#/+9',1"#;+('5)1&1"#)%9#
9'5212+%1
A
5
Thursday, 23 July 15
C@+0@+,)
4+3%9#L<#0)0'(1#E2&C#HW#5+/0)(21+%1N#
)M X+%'#+;#&C'#0)0'(1#;+3%9#&C)&#5+/0,'D2&.#
C',0'9#)553()5.
FM :+/0,'D2&.#2%5(')1'9#'((+(#F.#<WY#+%#)8'()6'#
)5(+11#0)0'(1
Z-2/0,'#8'(131#5+/0,'D#;+('5)1*%6N#[C'#'829'%5'\#
E)1#03F,21C'9#2%#JBR#2%#<?@AG#
###############P3'#+3&#2%#0(2%&#2%#B3631&G
]
6
Thursday, 23 July 15
:>--("D$&%$#=@0#+'#$&+$(''>"('D$&%$%&"#'().)$
%"&-$'&-12#3$=)7$)@-12#$-#./&0)
W
7
Thursday, 23 July 15
:@-12@'@.DE$!$'/#'F2@).
I
8
Thursday, 23 July 15
J#-&+)."(*&+E$KL#'.$&%$)@-12@'@.D$=)7$
'&-12#3@.D$@+$'2@-(.#$%&"#'()*+,
$a::#E)(/2%6#),)(/21&1#9+#not#;+('5)1&"#
&C'.#5(')&'#Z15'%)(2+1\#82)#5+/03&'(#12/3,)*+%1
@G-5'%)(2+1#)('N
)G -&+(2'1#)F+3&#ZEC)&#C)00'%'9#2%#&C'#;3&3('\
FG b2)1'9"#1+#&C'.#9+#%+&#0(+829'#8),29#;+('5)1&1#
Jc('6+(.#d#P3()%"#<??@MG
<G[C'#1&+(2'1#)('#F)1'9#+%#'D0'(&#e396/'%&1G#
B55+(92%6#&+#0(2+(#('1')(5C"#'D0'(&#e396/'%&1#
)F+3&#EC)&#E2,,#C)00'%#2%#complex,*uncertain*
situa:ons*)('#31','11N
)G -''(K135O'(#[C'+(.
FG ['&,+5Of1#<?K.')(#'D0'(2/'%&
H
9
Thursday, 23 July 15
MN66$1"&'#))$&%$'"#(*+,$'2@-(.#$)'#+("@&)$=@($
'&-1>.#"$)@->2(*&+)$@)$enormously*complex
@G >396/'%&1#)('#/)9'#+%#EC)&#8)(2)F,'1#&+#
2%5,39'#J'G6G#:g2M"#)%9#'D5,39'#J'G6G#-3%Mh#
<G >396/'%&1#)('#/)9'#+%#&C'#8),3'1#+;#/)%.#
0)()/'&'(1"#)%9#&C'2(#J%+%,2%')(M#(',)*+%1C201
LG B(+3%9#A?"???#6(29#1V3)('1#)('#/+9','9
=G c(29#1V3)('#/+9',1#2%&'()5&
AG `+9',1#)('#)9e31&'9#&+#0(+935'#'D0'5&'9#
+3&03&1
]G b396'&#;+(#5+/03&'(#12/3,)*+%1#21#'%+(/+31
@?
10
Thursday, 23 July 15
8"##+O$!"-)."&+,O$P$:&&+$QRGGST$+&H'/(+,#$
-&0#2$@)$sophis0catedly*simple
@G b)1'9#+%#)%#'D)/2%)*+%#+;#928'(1'#,+%6#
&'/0'()&3('#C21&+(2'1S
<G X+#,+%6K&'(/#&('%9
LG 7'8'(1),1#+%#),,#*/'#15),'1
=G :+((',)&'9#E2&C#1+,)(#5.5,'1#)%9#8)(2)*+%1#2%#
)5*82&.#
AG Q')O,.#5+((',)&'9#E2&C#:g2S
F3&#&'/0'()&3('#5C)%6'1#0('5'9'#:g2#5C)%6'1"#
)%9#C26C#:g2#,'8',1#C)8'#F''%#)11+52)&'9#E2&C#
25'#)6'1G
@@
11
Thursday, 23 July 15
:@-12#$C&"#'()*+,$6/#'F2@).$"(*+,)E
MN66$1"&U#'*&+)$=)7$+&H'/(+,#$%&"#'().)$
@<
12
Thursday, 23 July 15
K=@0#+'#$&+$(''>"('D$&%$MN66$1"&U#'*&+)$=)7$
+&H'/(+,#$%&"#'().)$>)@+,$V(02#D$0(.($
['1&1#+;#;+('5)1&1#+8'(#&C'#@IA@K@HWA#;+('5)1*%6#0'(2+9#.2',9'9#AI#
;+('5)1&1#;+(#C+(2R+%1#+;#H@#&+#@??#.')(1G#[C'#'((+(1#+;#&C'1'#$a::#
;+('5)1&1#E'('#@<G]#*/'1#,)(6'(#&C)%#&C+1'#;(+/#&C'#')12,.#3%9'(1&++9#
%+K5C)%6'#/+9',#Jc(''%"#B(/1&(+%6"#d#-++%#<??HMG
@L
:C)(&#;(+/#4+('5)1*%6#c,+F),#:,2/)&'#:C)%6'#J<?@=M
13
Thursday, 23 July 15
@=
4(+/#4+('5)1*%6#c,+F),#:,2/)&'#:C)%6'#J<?@=M
K=@0#+'#$&+$(''>"('D$&%$MN66$1"&U#'*&+)$=)7$+&H
'/(+,#$%&"#'().)$>)@+,$W&#/2#$P$X'6>22&'/$QRGGYT$
0(.($
14
Thursday, 23 July 15
Z(2@0(*&+$&=#"$0@L#"#+.$*-#H1#"@&0)[$0(.(E$
:@-@2("$"#)>2.)
4(+/#4+('5)1*%6#c,+F),#:,2/)&'#:C)%6'#J<?@=M
@A
15
Thursday, 23 July 15
\/#$6&-12#3@.D$N#+(2.D
$%#&C'#,)&'#@HW?"#)#('1')(5C#('82'E#;+3%9#&C)&#
5+/0,'D#;+('5)1*%6#/'&C+91#C)(/'9#)553()5.G
$%#&C'#,)&'#@HH?1"#&C'#4+('5)1*%6#a(2%520,'1#a(+e'5&#
9'8',+0'9#@LH#0(2%520,'1#;+(#;+('5)1*%6G#B,,#E'('#
12/0,'#&+#3%9'(1&)%9G
g3(#('5'%&#1.1&'/)*5#('82'E#;)2,'9#&+#T%9#)#12%6,'#
1&39.#1C+E2%6#&C)&#)#5+/0,'D#/'&C+9#E)1#/+('#
)553()&'#&C)%#)#12/0,'(#/'&C+9G
@]
16
Thursday, 23 July 15
\A&$]>#)*&+)$%&"$D&>
QC)&#0'(5'%&)6'#+;#0)0'(1#)&#&C21#
5+%;'('%5'#0(+0+1'#5+/0,'D#/'&C+91^
QC.#9+'1#&C)&#C)00'%^
#
@W
17
Thursday, 23 July 15
:#0>'#0$^D$'&-12#3@.D
-+/'#'829'%5'#1366'1&1#&C)&#&C'#0+03,)(2&.#+;#
5+/0,'D2&.#/).#F'#93'#&+#2%5'%*8'1N#
J@M('1')(5C'(1#)('#('E)(9'9#;+(#03F,21C2%6#2%#C26C,.#
()%O'9#e+3(%),1"#EC25C#;)8+(#5+/0,'D2&.h
J<M;+('5)1&'(1#5)%#31'#5+/0,'D#/'&C+91#&+#0(+829'#
;+('5)1&1#&C)&#1300+(&#9'5212+%K/)O'(1f#0,)%1h#)%9#
JLM;+('5)1&'(1f#5,2'%&1#/).#F'#2/0('11'9#)%9#
(')113('9#F.#2%5+/0('C'%12F2,2&.G#
@I
18
Thursday, 23 July 15
6&+'2>)@&+)
@G#[C+1'#EC+#0(';'(#&C'2(#;+('5)1&1#&+#F'#
)553()&'#1C+3,9#)55'0&#;+('5)1&1#+%,.#;(+/#
12/0,'#'829'%5'KF)1'9#0(+5'93('1G
<G#B,)(/2%6#$a::#&'/0'()&3('#0(+e'5*+%1#)('#
F)1'9#+%#0(+5'93('1#&C)&#)('#&++#5+/0,'D#&+#F'#
&(31&'9G
LG#7)&'#&C'#12/0,252&.#+;#;+('5)1*%6#0(+5'93('1#
31'9#;+(#0(+F,'/1#.+3#)('#2%&'('1&'9#2%#312%6#&C'#
V3'1*+%%)2('#)&S#
12/0,'K;+('5)1*%6G5+/G#
@H
19
Thursday, 23 July 15
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.