B
Ben Hur
Hi,
I wish to forecast values against a time series (1-120 months) and
using existing data as a "trendline". Is this possible?
For example I have months 1-120 in row 1 and data of an existing curve
in row 2. In rows 3-60 I have incomplete data (usually 60 months
worth) and I wish to extrapolate this data using a forecast function.
I cannot use the linear trend functions within Excel as the data often
reaches a maximum limit (different for each data set) at around month
75-85 and the linear trendline predicts values above this point.
What I am hoping for (I think) is some sort of forecasting function
which allows higher weighting of the most recent variable (i.e. month
84) rather than the earliest variable (month 1). I know how to do
this for smoothing methods (weighted moving average), but not as a
form of prediction. Can anyone help?
Example data:
1. Month 1 2 3 4 ..... 84 85 .... 120
2. Data1 2.2 3.1 3.2 3.4 6.6 6.7 6.7
3 Data2 1.4 1.5 2.7 3.1 5.9 ? ?
4. Data 3 6.9 7.2 7.3 7.8 10.4 ? ?
Thanks
BH
I wish to forecast values against a time series (1-120 months) and
using existing data as a "trendline". Is this possible?
For example I have months 1-120 in row 1 and data of an existing curve
in row 2. In rows 3-60 I have incomplete data (usually 60 months
worth) and I wish to extrapolate this data using a forecast function.
I cannot use the linear trend functions within Excel as the data often
reaches a maximum limit (different for each data set) at around month
75-85 and the linear trendline predicts values above this point.
What I am hoping for (I think) is some sort of forecasting function
which allows higher weighting of the most recent variable (i.e. month
84) rather than the earliest variable (month 1). I know how to do
this for smoothing methods (weighted moving average), but not as a
form of prediction. Can anyone help?
Example data:
1. Month 1 2 3 4 ..... 84 85 .... 120
2. Data1 2.2 3.1 3.2 3.4 6.6 6.7 6.7
3 Data2 1.4 1.5 2.7 3.1 5.9 ? ?
4. Data 3 6.9 7.2 7.3 7.8 10.4 ? ?
Thanks
BH