Автор: Пользователь скрыл имя, 17 Января 2011 в 16:35, курсовая работа
Цель данной работы заключается в построении прогноза по статистическим данным индустрии гостеприимства собранным за несколько предыдущих лет и анализ прогноза на будущий период.
Задачи данной работы могут быть сформулированы следующим образом: раскрытие понятия о временных рядах и существующих в индустрии гостеприимства методах построения прогнозов; приведение конкретного примера с помощью программы Statgraphics Plus - анализ данных по ежемесячной загрузке гостиниц Северной Ирландии, выявление трендов и моделей сезонности, анализ случайности; построение прогноза с помощью функции автоматическое прогнозирование и анализ полученных данных с их дальнейшей трактовкой и выработкой конкретных рекомендаций и выводов по данной ситуации.
Введение…………………………………………………………….……………3
I. Теоретическое обоснование прогнозирования в индустрии гостеприимства и туризма
1.Сущность и методы прогнозирования…………………………….…….….5
2.Понятие временных рядов и основные этапы их анализа……………....…7
3.Общая характеристика STATGRAPHICS и его особенности………….....10
II. Анализ временных рядов в STATGRAPHICS…………………………..12
III. Автоматическое прогнозирование временных рядов………………...22
Заключение………………………………………………………………….…..31
Список использованной литературы………………………………………..32
Приложения………………………………………………………………….….33
Expected number of runs = 40,975
Large sample test statistic z = 2,14482
P-value
= 0,0319669
Runs up and down
---------------------------
Number of runs up and down = 65
Expected number of runs = 55,0
Large sample test statistic z = 2,50058
P-value
= 0,0123991
Box-Pierce Test
---------------
Test based on first 24 autocorrelations
Large sample test statistic = 34,231
P-value
= 0,0806472
The StatAdvisor
---------------
Three tests have been run to determine whether or not adjusted
Occupancy rate is a random sequence of numbers. A time series of
random numbers is often called white noise, since it contains equal
contributions at many frequencies. The first test counts the number
of times the sequence was above or below the median. The number of
such runs equals 51, as compared to an expected value of 40,975 if the
sequence were random. Since the P-value for this test is less than
0.01, we can reject the hypothesis that the series is random at the
99% confidence level. The second test counts the number of times the
sequence rose or fell. The number of such runs equals 65, as compared
to an expected value of 55,0 if the sequence were random. Since the
P-value for this test is less than 0.01, we can reject the hypothesis
that the series is random at the 99% confidence level. The third test
is based on the sum of squares of the first 24 autocorrelation
coefficients. Since the P-value for this test is less than 0.10, we
can reject the hypothesis that the series is random at the 90%
confidence level. Since the three tests are sensitive to different
types of departures from random behavior, failure to pass any test
suggests that the time series
may not be completely random.
Automatic Forecasting - Occupancy rate
Seasonal Decomposition - Occupancy
rate
Analysis Summary
Data variable: Occupancy rate
Number of observations = 84
Start index = 1.97
Sampling interval = 1,0 month(s)
Length of seasonality = 12
Seasonal Decomposition
----------------------
Method: Multiplicative
The StatAdvisor
---------------
This procedure applies a multiplicative seasonal decomposition to
Occupancy rate. The purpose of the decomposition is to separate
Occupancy rate into trend-cycle, seasonal, and random components. The
data cover 84 time periods. Each of the tables and graphs shows
different aspects of the decomposition.
Data Table for Occupancy rate
Seasonal decomposition method:
Multiplicative
Period Data Trend-Cycle Seasonality Irregular Adjusted
------------------------------
1.97
29,0
2.97
36,0
3.97
38,0
4.97
39,0
5.97
45,0
6.97
49,0
7.97 40,0 42,25 94,6746 89,5224 37,8232
8.97 55,0 42,4583 129,539 102,651 43,5839
9.97 53,0 42,5833 124,462 102,939 43,835
10.97 47,0 42,625 110,264 107,017 45,6159
11.97 42,0 42,7083 98,3415 105,722 45,1519
12.97 33,0 42,5833 77,4951 108,084 46,0258
1.98 31,0 42,5417 72,8697 103,035 43,8329
2.98 39,0 42,5417 91,6748 103,663 44,1002
3.98 38,0 42,25 89,9408 100,089 42,2875
4.98 40,0 41,875 95,5224 97,5726 40,8585
5.98 46,0 41,5 110,843 98,4212 40,8448
6.98 45,0 41,1667 109,312 91,206 37,5465
7.98 43,0 40,9583 104,985 99,2715 40,6599
8.98 52,0 40,75 127,607 101,12 41,2066
9.98 49,0 40,5417 120,863 99,963 40,5267
10.98 42,0 40,5417 103,597 100,546 40,7632
11.98 38,0 40,625 93,5385 100,558 40,8517
12.98 29,0 40,9583 70,8037 98,7514 40,4469
1.99 30,0 41,4583 72,3618 102,317 42,4189
2.99 35,0 41,75 83,8323 94,7954 39,5771
3.99 37,0 42,0417 88,0079 97,9378 41,1747
4.99 41,0 42,4167 96,6601 98,7348 41,88
5.99 47,0 42,75 109,942 97,6204 41,7327
6.99 52,0 42,9583 121,048 100,998 43,387
7.99 48,0 42,9167 111,845 105,758 45,3878
8.99 54,0 42,875 125,948 99,8051 42,7915
9.99 54,0 42,9167 125,825 104,067 44,662
10.99 46,0 43,0833 106,77 103,626 44,6454
11.99 42,0 43,375 96,83 104,097 45,1519
12.99 30,0 43,5 68,9655 96,1877 41,8417
1.00 28,0 43,4167 64,4914 91,1885 39,591
2.00 36,0 43,25 83,237 94,1222 40,7078
3.00 37,0 43,0417 85,9632 95,6624 41,1747
4.00 45,0 42,7083 105,366 107,627 45,9659
5.00 50,0 42,3333 118,11 104,874 44,3965
6.00 52,0 42,2083 123,198 102,793 43,387
7.00 46,0 42,3333 108,661 102,748 43,4967
8.00 52,0 42,5 122,353 96,9567 41,2066
9.00 51,0 42,625 119,648 98,9579 42,1808
10.00 41,0 42,4167 96,6601 93,8136 39,7926
11.00 38,0 42,0 90,4762 97,266 40,8517
12.00 31,0 41,875 74,0299 103,251 43,2364
1.01 30,0 41,875 71,6418 101,299 42,4189
2.01 38,0 41,9583 90,566 102,41 42,9694
3.01 38,0 42,0417 90,3865 100,585 42,2875
4.01 39,0 42,0833 92,6733 94,6624 39,8371
5.01 46,0 42,2083 108,983 96,7695 40,8448
6.01 53,0 42,25 125,444 104,666 44,2214
7.01 45,0 42,2917 106,404 100,613 42,5511
8.01 55,0 42,375 129,794 102,853 43,5839
9.01 50,0 42,5 117,647 97,3029 41,3537
10.01 43,0 42,7083 100,683 97,7179 41,7337
11.01 39,0 43,0 90,6977 97,5041 41,9268
12.01 31,0 43,0833 71,9536 100,355 43,2364
1.02 31,0 43,0417 72,0232 101,838 43,8329
2.02 39,0 42,9583 90,7856 102,658 44,1002
3.02 40,0 42,8333 93,3852 103,922 44,5132
4.02 42,0 42,8333 98,0545 100,159 42,9015
5.02 50,0 42,7917 116,845 103,75 44,3965
6.02 51,0 42,6667 119,531 99,7328 42,5527
7.02 46,0 42,5833 108,023 102,145 43,4967
8.02 52,0 42,625 121,994 96,6723 41,2066
9.02 50,0 42,7083 117,073 96,8283 41,3537
10.02 43,0 42,875 100,292 97,3381 41,7337
11.02 38,0 43,0417 88,2865 94,912 40,8517
12.02 29,0 43,2917 66,9875 93,4289 40,4469
1.03 31,0 43,6667 70,9924 100,381 43,8329
2.03 40,0 44,1667 90,566 102,41 45,2309
3.03 41,0 44,7917 91,5349 101,863 45,626
4.03 45,0 45,375 99,1736 101,302 45,9659
5.03 51,0 45,9167 111,071 98,6231 45,2845
6.03 56,0 46,4167 120,646 100,663 46,7245
7.03
50,0
8.03
60,0
9.03
57,0
10.03
50,0
11.03
44,0
12.03
35,0
The StatAdvisor
---------------
This table shows each step of the seasonal decomposition. The
trend-cycle column shows the results of a centered moving average of
length 12 applied to Occupancy rate. The seasonality column shows the
data divided by the moving average and multiplied by 100. Seasonal
indices are then computed for each month by averaging the ratios
across all observations in that month, and scaling the indices so that
an average month equals 100. The data is then divided by the
trend-cycle and seasonal estimates to give the irregular or residual
component. This component
is then multiplied by 100.
Seasonal Indices for Occupancy
rate
Seasonal decomposition method:
Multiplicative
Season Index
------------------------
1 70,7232
2 88,435
3 89,861
4 97,8988
5 112,621
6 119,852
7 105,755
8 126,193
9 120,908
10 103,034
11 93,0193
12
71,6989
The StatAdvisor
---------------
This table shows the seasonal indices for each month, scaled so
that an average month equals 100. The indices range from a low of
70,7232 in month 1 to a high of 126,193 in month 8. This indicates
that there is a seasonal swing from 70,7232% of average to 126,193% of
average throughout the course
of one complete cycle.
Analysis Summary
Data variable: Occupancy rate
Number of observations = 84
Start index = 1.97
Sampling interval = 1,0 month(s)
Length of seasonality = 12
Forecast Summary
----------------
Seasonal adjustment: Multiplicative
Forecast model selected: Simple exponential smoothing with alpha = 0,7043
Number of forecasts generated: 24
Number of periods withheld
for validation: 0
Информация о работе Прогнозирование в индустрии гостеприимства и туризма