Year Quarter Sales Smoothed Levels Standard Errors

2021 1 165000 #N/A #N/A

2 253000 165000 #N/A
3 316000 226600 #N/A
4 287000 289180 #N/A
2022 1 257000 287654 72436.4374239
2 308000 266196.2 54579.544752
3 376000 295458.86 29955.3991263
4 351000 351837.658 55299.4895785

Exponential Smoothing Methods are a family of forecasting models. They use weighted averages of pa
Standard Errors (SE) measure the uncertainty or variability in the forecast errors.
Smoothed Levels refer to the estimated or "smoothed" values of a time series after eliminating shor

Exponential Smoothing
400000
f(x) = 185.41666666654 x⁶ − 5683.4935897399 x⁵ + 66706.730769189 x⁴ − 374909.01806503 x³ +
350000
R² = 0.999586624125678
300000
250000
Value

200000
150000
100000
50000
0
1 2 3 4 5 6
Data Point
 Forecast (exponential) Forecast(linear)
224571.720494534 230700
261144.595632658
303673.586674082
269650
308600
Exponent
353128.683441032 347550 400000
224571.720494534 230700

Value
200000
261144.595632658 269650
303673.586674082 308600 0
353128.683441032 347550 1 2 3
Data

models. They use weighted averages of past observations to forecast new values. The idea is to give more imp
ty in the forecast errors.
alues of a time series after eliminating short-term fluctuations, making it easier to observe long-term trends or p

xponential Smoothing

x⁵ + 66706.730769189 x⁴ − 374909.01806503 x³ + 1026075.5099061 x² − 1202439.8310013 x + 655124.99999952

Actual
Polynomial (Actual)

4 5 6 7 8
Data Point
 Exponential Smoothing
400000
Actual
Value

200000
Forecast
0
1 2 3 4 5 6 7 8
Data Point

The idea is to give more importance to recent values in the series. Thus, as observations get older in time, the

observe long-term trends or patterns.

Exponential Smoothing
655124.99999952 400000
350000 f(x) = 22178.5714285714 x + 189321.428571429
300000 R² = 0.684018964608975
250000
ctual
Value

200000
olynomial (Actual)
150000
100000
50000
0
1 2 3 4 5 6 7
Data Point
ons get older in time, the importance of these values get exponentially smaller.

ial Smoothing

189321.428571429

Actual
Linear (Actual)

5 6 7 8
t
Period Sales Forecast.Linear
1 20
2 32
3 51 Linea
4 43 140
5 62 120
6 63
100
7 82
8 75 80
9 92 60
10 89 89
40
11 103.53
12 111.28 20
13 119.04 0
0 2 4 6

Sales
 FORECAST.LINEAR is a function in Exce
Linear Trend

2 4 6 8 10 12 14

Sales Forecast.Linear
NEAR is a function in Excel that provides a forecast using linear regression.
Period Visitors
1 74
2 59
3 51
4 42 Seasona
5 83 100
6 67 90
7 54 80
8 44 70
60
9 87
50
10 68
40
11 55 30
12 49 49.00 20
13 90.42 10
14 72.37 0
0 2 4 6 8
15 60.34
16 52.72 Column A C
 FORECAST.ETS is a function in Excel u

Seasonal Pattern

4 6 8 10 12 14 16 18

Column A Column B Column C

TS is a function in Excel used for Exponential Smoothing-based forecasting, particularly for time series data
ularly for time series data. It allows users to generate forecasts by accounting for seasonality, trends, and error
onality, trends, and error smoothing in the data. The method behind this function is based on ETS (Exponentia
sed on ETS (Exponential Triple Smoothing), which can model both seasonality and trends in data, making i
d trends in data, making it a popular choice for forecasting in various fields like sales, inventory management, a
nventory management, and finance.
Period Visitors Forecast( Visitors) Lower Confidence Bound( Visitors)
1 74
2 59
3 51
4 42
5 83
6 67
7 54
8 44
9 87
10 68
11 55
12 49 49 49.00
13 90.4232497483681 86.07
14 72.539153905247 67.67
15 61.9348352190018 56.60
16 52.6988324175824 46.93

13 90 86
14 73 68
15 62 57
16 53 47
Upper Confidence Bound( Visitors)

100

90

80

70

60
49.00
94.78
77.41 50
67.27
58.47
40

95
77 30
67
58 20

10

0
1 2 3 4 5
Visitors Forecas
3 4 5 6 7 8 9 10 11 12 13
Visitors Forecast( Visitors) Lower Confidence Bound( Visitors) Upper Confidence Bound( Visitors)
 12 13 14 15 16
nfidence Bound( Visitors)
Period Visitors
1 74
2 59
3 51
4 42
5 83
6 67
7 54
8 44
9 87
10 68
11 55
12 49