Environmental Engineering Reference
In-Depth Information
EðS
t
Þ
¼
S
0
e
k
S
ðtt
0
Þ
þ S
m
ð
e
k
S
ðtt
0
Þ
Þ
;
1
ð
52
Þ
which amounts to
EðS
tþ
D
t
Þ
¼
S
t
e
k
S
D
t
e
k
S
D
t
þ S
m
ð
1
Þ
:
ð
53
Þ
Also, the conditional variance is
r
S
2
k
S
½
e
2
k
S
ðtt
0
Þ
:
Var
ðS
t
Þ
¼
1
ð
54
Þ
Since both mean and variance remain
nite as
t !1
, this process is stationary.
Equation (
11
) is the continuous-time version of a
rst-order autoregressive
process AR (
1
) in discrete time:
e
k
D
t
ÞþS
t
e
k
D
t
S
tþ
D
t
¼
S
m
ð
1
þ e
tþ
D
t
¼
a þ bS
t
þ e
tþ
D
t
;
ð
55
Þ
where
e
t
:
Nð
0
;
r
e
Þ
, and the following notation holds:
a
a S
m
ð
1
bÞ)S
m
¼
ð
56
Þ
b
;
1
) k
¼
ln
b
b e
kDt
ð
57
Þ
D
t
:
Also
r
2
2
k
½
2
e
2
k
D
t
ðr
e
Þ
¼
1
)
ð
58
Þ
2
2
2
kðr
e
Þ
ðr
e
Þ
ln
b
2
) r
2
¼
e
2
kDt
¼
:
1
D
t
½
b
2
1
58
) enable the continuous-time process parameters (
k
,
S
m
,
r
)to
be recovered on estimating the regression coef
Equations (
56
-
cients (
a
,
b
) and the standard
deviation of the regression residuals (
r
e
).
References
1. Abadie LM, Chamorro JM (2013) Investment in energy assets under uncertainty. Springer,
London
2. Abadie LM, Ortiz RA, Galarraga I (2012) Determinants of energy ef
ciency investments in
the U.S. Energy Policy 45:551
566
3. Abadie LM (2009) Valuation of long-term investments in energy assets under uncertainty.
Energies 2(3):738
-
768
-
Search WWH ::
Custom Search