on 17-May-2015 (Sun)

Flashcard 150891134

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
In simple exponential smoothing satisfying the formula:
$$\hat{x}_{n+1}$$ = αxn + (1 − α)$$\hat{x}_n$$, the lower the value of α, the [smoother or rougher?] the forecasts will be because they are not aﬀected much by recent values.
smoother

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
In simple exponential smoothing satisfying the formula: \hat{x}_{n+1} = αx n + (1 − α)\hat{x}_n, the lower the value of α, the smoother the forecasts will be because they are not aﬀected much by recent values.

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 150891166

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
The 1-step ahead forecast error at time t, which is denoted et, is the diﬀerence between the observed value and the 1-step ahead forecast of Xt:
et = xt - $$\hat{x}_t$$
The sum of squared errors, or SSE, is given by
SSE = [...]
Given observed values x1 ,x2 ,...,xn ,the optimal value of the smoothing parameter α for simple exponential smoothing is the value that minimizes the sum of squared errors.
$$\large SSE = \sum_{t=1}^ne_t^2 = \sum_{t=1}^n(x_t-\hat{x}_t)^2$$

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
step ahead forecast error at time t, which is denoted e t , is the diﬀerence between the observed value and the 1-step ahead forecast of X t : e t = x t - $$\hat{x}_t$$ The sum of squared errors, or SSE, is given by SSE <span>= $$\large \sum_{t=t}^ne_t^2 = \sum_{t=t}^n(x_t-\hat{x}_t)^2$$ Given observed values x 1 ,x 2 ,...,x n ,the optimal value of the smoothing parameter α for simple exponential smoothing is the value that minimizes the sum of squared errors.<

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 150891282

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
Give a formula for additive time series model with constant level and no seasonality
Xt = m + Wt

status measured difficulty not learned 37% [default] 0

pdf

cannot see any pdfs

Annotation 150891291

 #m249 #mathematics #open-university #statistics #time-series Suppose that the time series Xt can be described by an additive non-seasonal model with a linear trend component, that is, Xt = m + bt + Wt , where b is the slope of the trend component mt = m + bt. Note that Xt+1 = m + b(t +1) + Wt+1 =(m + bt) + b + Wt+1 = mt + b + Wt+1

pdf

cannot see any pdfs

Flashcard 150891298

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
Suppose that the time series Xt can be described by an additive non-seasonal model with a linear trend component, that is,
Xt = [...] , where b is the slope of the trend component mt = m + bt.
Note that
Xt+1 = m + b(t +1) + Wt+1
=(m + bt) + b + Wt+1
= mt + b + Wt+1
m + bt + Wt

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
Suppose that the time series X t can be described by an additive non-seasonal model with a linear trend component, that is, X t = m + b t + W t , where b is the slope of the trend component m t = m + bt. Note that X t+1 = m + b(t +1) + W t+1 =(m + bt) + b + W t+1 = m t + b + W t+1 <

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 150891307

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
Suppose that the time series Xt can be described by an additive non-seasonal model with a linear trend component, that is,
Xt = m + bt + Wt , where b is the [...] of the trend component mt = m + bt.
Note that
Xt+1 = m + b(t +1) + Wt+1
=(m + bt) + b + Wt+1
= mt + b + Wt+1
slope

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
Suppose that the time series X t can be described by an additive non-seasonal model with a linear trend component, that is, X t = m + b t + W t , where b is the slope of the trend component m t = m + bt. Note that X t+1 = m + b(t +1) + W t+1 =(m + bt) + b + W t+1 = m t + b + W t+1

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 150891319

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
Suppose that the time series Xt can be described by an additive non-seasonal model with a linear trend component, that is,
Xt = m + bt + Wt , where b is the slope of the trend component mt = [...].
Note that
Xt+1 = m + b(t +1) + Wt+1
=(m + bt) + b + Wt+1
= mt + b + Wt+1
m + bt

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
Suppose that the time series X t can be described by an additive non-seasonal model with a linear trend component, that is, X t = m + bt + W t , where b is the slope of the trend component m t = m + bt. Note that X t+1 = m + b(t +1) + W t+1 =(m + bt) + b + W t+1 = m t + b + W t+1

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 150891325

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
Suppose that the time series Xt can be described by an additive non-seasonal model with a linear trend component, that is,
Xt = m + bt + Wt , where b is the slope of the trend component mt = m + bt.
Note that
Xt+1 = m + b[...] + Wt+1
=(m + bt) + b + Wt+1
= mt + b + Wt+1
(t +1)

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
pose that the time series X t can be described by an additive non-seasonal model with a linear trend component, that is, X t = m + bt + W t , where b is the slope of the trend component m t = m + bt. Note that X t+1 = m + b<span>(t +1) + W t+1 =(m + bt) + b + W t+1 = m t + b + W t+1 <span><body><html>

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 150891331

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
Suppose that the time series Xt can be described by an additive non-seasonal model with a linear trend component, that is,
Xt = m + bt + Wt , where b is the slope of the trend component mt = m + bt.
Note that
Xt+1 = m + b(t +1) + Wt+1
=[...]
= mt + b + Wt+1
(m + bt) + b + Wt+1

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
can be described by an additive non-seasonal model with a linear trend component, that is, X t = m + bt + W t , where b is the slope of the trend component m t = m + bt. Note that X t+1 = m + b(t +1) + W t+1 =<span>(m + bt) + b + W t+1 = m t + b + W t+1 <span><body><html>

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 150891340

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
Suppose that the time series Xt can be described by an additive non-seasonal model with a linear trend component, that is,
Xt = m + bt + Wt , where b is the slope of the trend component mt = m + bt.
Note that
Xt+1 = m + b(t +1) + Wt+1
=(m + bt) + b + Wt+1
= [...]
mt + b + Wt+1

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
n-seasonal model with a linear trend component, that is, X t = m + bt + W t , where b is the slope of the trend component m t = m + bt. Note that X t+1 = m + b(t +1) + W t+1 =(m + bt) + b + W t+1 = <span>m t + b + W t+1 <span><body><html>

Original toplevel document (pdf)

cannot see any pdfs

Annotation 150891346

 #m249 #mathematics #open-university #statistics #time-series simple exponential smoothing: the term exponential refers to the fact that the weights α(1 − α)i lie on an exponential curve.

pdf

cannot see any pdfs

Flashcard 150891353

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
simple exponential smoothing: the term exponential refers to the fact that the weights [give formula] lie on an exponential curve.
α(1 − α)i

status measured difficulty not learned 37% [default] 0

Parent (intermediate) annotation

Open it
simple exponential smoothing: the term exponential refers to the fact that the weights α(1 − α) i lie on an exponential curve.

Original toplevel document (pdf)

cannot see any pdfs

Flashcard 150891359

Tags
#m249 #mathematics #open-university #statistics #time-series
Question
what does the expanded m-times (i.e. non recursive) simple exponential smoothing formula looks like?
fully recursive is:

$$\hat{x}_{n+1}$$= αxn + (1 − α)$$\hat{x}_n$$
$$\large \hat{x}_{n+1} = \sum_{i=0}^m\alpha(1-\alpha)^ix_{n-i}+(1-\alpha)^{m+1}\hat{x}_{n-m}$$
If a time series X t is described by an additive model with constant level and no seasonality, 1-step ahead forecasts may be obtained by simple exponential smoothing using the formula $$\hat{x}_{n+1}$$= αx n + (1 − α)$$\hat{x}_n$$ where: x n is the observed value at time n, $$\hat{x}_n$$​and $$\hat{x}_{n+1}$$are the 1-step ahead forecasts of X n and X n+1 , and α is a smoothing parameter, 0 ≤ α ≤ 1.