Analysis Demand Of Ecotourism In Bakelalan Tourism

Long Ritan, Long Rusu, Buduk Bui, Long Rangat, Buduk Aru, and Pa'Tawing.

The name of Ba'Kelalan is derived from the Kelalan River and Ba ' which brings the significance moisture land in the Lun Bawang languages. Its population is about 1500 nevertheless it is home to 8000 peoples as they have fallen in love with the topographic point and be given to come and see this topographic point many times. In the cool mountain clime which is an norm of 24A°C, temperate fruits such as apples, Citrus reticulata oranges, strawberry, grape, and vanilla are grown. Ba'Kelalan besides produces rice and mountain salt is obtained from the nearby hills.

Tourist reaching to this topographic point has increased in recent old ages. Nowadays, Ba'Kelalan is available with a 9-hole golf class, and the colony is besides the easiest point to entree for sing Kayan Mentarang National Park ( KMNP ) which is located in Krayan, East-Kaliamantan, Indonesia. Ba'Kelalan itself is located near the country of Pulong Tau National Park ( PTNP ) .

The chief activity of the people of Ba'Kelalan is farming. They works moisture Paddy as their primary agriculture activities. However, Ba'Kelalan is good known as the lone topographic point in Sarawak which produces apples. Ba'Kelalan apples has become the icon of Ba'Kelalan and the the Lun Bawang who live in that country today.

What makes this topographic point go more familiar among the tourer is the civilization of the Lun Bawang people and the one-year Apple Feast which is held in May every twelvemonth. Another event is the Escaped of The Hearth of Borneo, 4WD, and the Home stay plan. These events attract more tourers to come and see this topographic point. Hence, this will assist to increase the income of the people in Ba'Kelalan.

Definition of Forecasting

Prediction is defined as the anticipation of future events based on known past values of relevant variables ( Makridakis, S. , Wheelright, S. C. & A ; Hyndman, R. J. , 1998 ) . In other words, prediction is conveying the significance of the procedure of analysing current and historical informations to find future tendencies. Normally, economic experts use calculating techniques in order to find future economic tendencies. Univariate Modelling Techniques are methods for analysing informations on a individual variable at a clip. Examples of Univariate Modelling Techniques are the NaA?ve Models, Methods of Average, the Exponential Smoothing Techniques and the Box-Jenkins Methodology. Both Double Exponential Smoothing and Holt 's Method illustrated in this survey are classified in the Exponential Smoothing Techniques.

Ecotourism Demand

Tourism is one of the major contributing sectors for Malaysia 's economic growing for several old ages. Year by twelvemonth, the figure of international tourer reachings to Malaysia shows an upward tendency which supports the state 's political stableness, aside from several plan has and bundles introduced by the Malayan authorities to promote international tourer to see Malaysia.

This subdivision explains the most relevant features on the demand of ecotourism in Ba'Kelalan, Sarawak. The volume, composing and recent development of tourer flows are investigated by utilizing the figure of reachings of tourers for the period between 2000-2010. The purpose of this work is to look into

the tourer demand which includes the international and local tourer. Based on the information collected it was observed that the reaching of international tourers from United Kingdom, Indonesia, Brunei, Australia and Hong Kong are ranked top in sing Ba'Kelalan during the forecasted clip period.

Table 1: Tourist reachings to Ba'Kelalan ( 2000-2010 )

Year

No. Of Tourist

1999

210

2000

207

2001

217

2002

289

2003

153

2004

64

2005

210

2006

369

2007

800

2008

1650

2009

2150

2010

2750

Beginnings: Visitor Books of Ba'Kelalan and Borneo Jungle Safari ( BJS )

Table 1 shows the tourer reachings to Ba'Kelalan during the period of 2000 until 2010, it can be seen that the figure of tourers rose annually from 2000 until 2010. In fact, tourers increased between 2000 and 2002 but there was a 2-year diminution is observed in 2003 and 2004. The world-wide SARS epidemic in twelvemonth 2003 may explicate the lessening of tourer reaching to the survey country.

Literature Review

Over the past 3 decennaries, we have seen many surveies in calculating international touristry demand done by both touristry research workers and practicians. Basically, the literature on patterning and calculating touristry demand is immense with assorted types of empirical analysis. Some of the research workers apply cross-sectional informations. However, most of them use the method of calculating to analyse touristry demand.

The increasing involvement in surveies on touristry demand has been motivated by the rapid growing of the touristry industry across the universe as the economic system tends to be stable in these decennaries. The last four decennaries have seen great developments in touristry demand analysis, in footings of the diverseness of research involvements, the deepness of theoretical foundations, and progresss in research methodological analysiss. Modeling of touristry demand in order to analyse the effects of assorted determiners, and accurate prediction of future touristry demand, are fundamentally two major focal points of touristry demand surveies.

A Time-series theoretical account as we know explains a variable with respect to its ain yesteryear and a random perturbation term. Particular attending should be paid to researching the historic tendencies and forms such as seasonality of the clip series involved, and to foretell the hereafter of this series based on the tendencies and forms identified in the theoretical account. Since clip series theoretical accounts merely require historical observations of a variable, it is less dearly-won in informations aggregation and theoretical account appraisal.

The major constructs and step of touristry demand or touristry visits can take topographic point for assorted grounds such as vacations, concern trips, visits to friends and relations ( VFR ) , conferences, and pilgrim's journeies and so on. In other words, the term 'tourism demand ' can be defined for a peculiar finish as the measure of the touristry merchandise ( Hong Kong Tourism Demand Forecasting System ) . It is refers to the combination of touristry goods and services that the consumer is willing to pay during the specific clip under the given set of conditions. The clip period can be a month, a one-fourth or a twelvemonth. Hence, the conditions that relate to the measure of touristry demanded is including the touristry monetary values for the finish such as the tourers ' cost of populating in their finish and travel costs to the finish. Besides that, the handiness of the

touristry monetary values for viing or utility finishs, possible consumers ' incomes, advertisement outgo, gustatory sensations of consumers in the beginning states, and other societal, cultural, geographical and political factors can be taken into history in order to mensurate the value of touristry demand.

It is due to being one of the of import countries in touristry research that touristry demand patterning and prediction has attracted much attending of both faculty members and practicians as mentioned before. Harmonizing to a comprehensive reappraisal by Li et Al ( 2005 ) in his diary, 420 surveies on these subjects were published during the period of 1960-2002. The bulks of these surveies focus on the application of different techniques, both qualitative and quantitative, to pattern and calculate the demand for touristry in assorted finishs. These surveies besides have attempted to set up prediction rules that could be used to steer the practicians in choosing prediction techniques. However, this attempt failed. As Witt and Song ( 2000 ) and Li et Al ( 2005 ) concluded that the public presentation of the prediction theoretical accounts varies harmonizing to the information frequences used in the theoretical account appraisal, the destination-origin state or part braces under consideration and the length of the prediction skylines concerned. There has non been a solution for touristry demand prediction.

Several Numberss of articles reappraisals on touristry demand calculating have been published over the last few decennaries. These include Crouch ( 1994 ) , Li et Al ( 2005 ) , Lim ( 1997a, 1997b and 1999 ) and Witt and Witt ( 1995 ) . These reappraisals cover the surveies published largely during the period of 1960-2000. Although a few surveies published between 2000 and 2004 are included in the latest reappraisal of Li et Al ( 2005 ) , the focal point of that reappraisal was on the econometric attack merely. This paper does non try to double the attempts made by old research workers. This paper will supply demand patterning and prediction, including clip series theoretical accounts utilizing the univariate mold such as naA?ve prognosis, mean alteration theoretical account, exponential smoothing and Holt 's Method which take into history to happen the smallest RMSE. This is as the consequence to happen which theoretical account is accurate to calculate the demand of touristry.

To measure the best theoretical account, it will take into history the comparing of consequences of each method through the possible standards such as the six common 1s are the average square mistake ( ME ) , the average absolute mistake ( MAE ) , the average per centum mistake ( MPE ) , the average absolute mistake ( MAE ) , the average square mistake ( MSE ) , and the root mean square mistake ( RMSE ) . Basically, ME and MPE are non frequently used as steps of prognosis truth because of big positive mistakes which can offsetted by big negative mistake. The other steps, MAE, MAPE, MSE, and RMSE are the best comparison option to compare the prediction theoretical accounts for a given series ( J. Holton, Barry Keating, John Galt 2009 ) . However,

usually we tend to concentrate on RMSE to measure the truth of assorted calculating method because it is easy for most people to construe due to its similarity to the basic statistical construct of standard divergence. Hence, it is one of the most normally used steps of prognosis truth.

Data and Methodology

This subdivision describes briefly on the statistical techniques applied to analyse the informations collected from the Visitor Book in Ba'Kelalan and Borneo Jungle Safari. Univariate Modelling Technique was applied to foretell the hereafter values of demand on ecotourism based on past observations given in the times series, by suiting a theoretical account to the informations.

The sum of tourer reaching to the survey country is collected from the twelvemonth 2000 until 2010. These informations were used to find the suited theoretical account to suit the information. Time series calculating analysis and prognosis theoretical accounts were applied to foretell the demand of ecotourism in Ba'Kelalan by utilizing these four types of prognosis theoretical accounts such as NaA?ve with Trend Mode, Average Change Model, Exponential Smoothing and Holt 's Method.

NaA?ve with Trend Model

This theoretical account implies that all future prognosis can be set to be the existent observed value in the most recent clip period. NaA?ve prognosiss assume that recent periods are the best forecasters of the hereafter ( John E. Hanke and Dean W.Wichern, 2009 ) . The value Yt/Yt-1 measures the tendency. If Yt is greater than Yt-1 so the tendency is upward and likewise if Yt is the existent value in clip t1 and Yt-1 is the existent value in the preceding clip period. The one measure in front to calculate is represented as,

( 1 )

Where:

Yt is the existent value in clip T, and

Yt-1 is the existent value in the preceding clip period.

This theoretical account is extremely sensitive to the alterations in the existent values. A crisp addition or sudden bead in the values will badly impact the prognosis. In add-on, suiting this theoretical account type will ensue in the loss of the first two observations in the series. Besides, this theoretical account is merely suited to be used for the short clip series.

Average Change Model

The mean alteration theoretical account or usually known as moving norm is based on the premiss that the prognosis value is equal to the existent value in the current period plus the norm of the absolute alterations experienced up to that point in clip. The one measure in front prognosis is given as ;

( 2 )

This theoretical account is utile when the historical informations that we are traveling to analyse are characterized by period to period alterations that are about of the same size. However, this theoretical account tends to dawdle behind turning points and that all the period is weighted every bit, irrespective of their importance, when deducing the prognosis value.

Equation below gives the simple moving mean prognosis. A moving norm of order K, MA ( K ) , is computed by ;

( 3 )

Where:

Ft+1 = the prognosis value for the following period

Crosstalk = the existent value at period T

K = the figure of footings in the moving norm

Exponential Smoothing

Simple exponential smoothing every bit

good as traveling mean utilizations merely the past values of a clip series to calculate future values of the same series and is decently employed when there is no tendency or seasonality nowadays in the information. The purpose is to gauge the current degree whereby this degree estimation is so used as the prognosis of future values. Exponential smoothing continually revises the estimation in the visible radiation of more-recent experiences. This method is based on averaging or smoothed the past values of a series in the diminishing exponential mode. This means that the most recent observation receives the largest weight, I± ( where 0 & lt ; I± & lt ; 1 ) , the following observation by ( 1 - I± ) 2I± , and so on. The figure we choose for I± is called the degree of smoothing changeless. Therefore,

( 4 )

More officially, the equation is as below ;

( 5 )

Where:

= Forecast value for period t+1

I± = Smoothing invariable ( 0 & lt ; I± & lt ; 1 )

Crosstalk = Actual value now ( in period T )

Ft = Forecast ( eg: smoothed ) value for period T

As a usher in taking I± , select value which is close to 0 if the series has a great trade of random fluctuation. If we wish the prognosis values to depend strongly on recent alterations in the existent values we should choose the value which is close to 1. The root mean squared mistake ( RMSE ) is frequently used as the standard for delegating an appropriate smoothing invariable ; the smoothing invariables which give the smallest RMSE would be selected as the theoretical account which provides the smallest mistake in bring forthing the extra prognosiss ( Wilson, Barry and John Galt 2009 ) . Basically, the little values of I± by and large work best when simple exponential smoothing is the most appropriate theoretical account.

Holt 's Method

This theoretical account is the extension of the smoothing theoretical accounts. Holt 's Method sometimes called dual exponential smoothing, is a technique to smooth the tendency and the incline straight by utilizing the different smoothing invariables. It besides provides more flexibleness in choosing the parametric quantity value which the tendency and the inclines are tracked. Three equations and two smoothing invariable are used in this theoretical account. The Holt 's Method equations are represented as follows:

Exponential smoothed series:

( 6 )

Trend estimation:

( 7 )

Therefore, the one measure in front prognosis is:

( 8 )

Where:

= Smoothed value for period t+1

I± = Smoothing invariable for the degree ( 0 & lt ; I± & lt ; 1 )

Crosstalk = Actual value now ( in period T )

Ft = Forecast ( eg. Smoothed ) value for the period T

= Trend estimation

I? = Smoothing invariable for the tendency estimation ( 0 & lt ; I? & lt ; 1 )

m = Number of periods in front to be forecast

= Holt 's prognosis value for period T + m

Analysis and Consequences

Univariate prediction theoretical account was used to foretell the demand of ecotourism for the first one-fourth of 2011. The appraisal was done utilizing the first one-fourth of 2000 to

the 4th one-fourth of 2010. Measure of Root Mean Square Error ( RMSE ) is used to mensurate the difference between the prognosis and the existent values in the station estimated period. ( Mentzer and Kahn, 1995 ; Witt and Witt 1992 ; Armstrong 2001 ) . We chose to compare the value of RMSE among the prediction theoretical account because measures the differences between values predicted by a theoretical account or an calculator and the values really observed from the point being modeled or estimated. Since the RMSE is a good step of truth, it is ideal if it is little.

The truth of the prediction theoretical accounts over figure periods is evaluated so that we can place which theoretical account by and large works the best. Among the standards which were taken into history were average mistake ( ME ) , the average absolute per centum mistake ( MAE ) , the average per centum mistake ( MPE ) , the average absolute mistake ( MAE ) , the average absolute per centum mistake ( MAPE ) , the average square mistake ( MSE ) , and the root mean square mistake ( RMSE ) .

Throughout this survey, we focused on root mean square mistake ( RMSE ) to measure the comparative truth of assorted calculating method. The RMSE is calculated as follows:

( 9 )

Where ;

At = Actual value in period T

Ft = Forecast value in period T

n = Number of periods used in the computation

Choice of Model

Table 3 represents the sum-ups and comparing on root mean square mistake ( RMSE ) figures for NaA?ve with Forecast, Average Change Model, Exponential Smoothing, and Holt 's Method. On the footing of the size of RMSE calculated over the clip period, it can be concluded that the most suited theoretical account to calculate the demand of ecotourism is Average Change Model since it has the smallest value of RMSE ( 29.85 ) compared to other prediction technique.

The Average Change Model is the best theoretical account to calculate this information because of several grounds. Average Change Model is used to calculate as it reduces the consequence of impermanent fluctuations in informations, better the tantrum of informations to a line whereby it is called a smoothing procedure to demo the information 's tendency more clearly, and highlight any value above or below the tendency.

Table 3: RMSE Valuess by type of theoretical account

Type of theoretical account

NaA?ve Forecast

Average Change Model

Exponential Smoothing

I±=0.48

Holt 's Method

I± =0.11, I?=1.00

RMSE

123.60

29.85

110.51

97.50

The first method which is NaA?ve with Trend Model is done utilizing Microsoft Excel whereby all the informations are analyzed through appraisal of the prognosis value. The simplest naA?ve prediction theoretical account, in which the prognosis value is equal to the old ascertained value, can be described in algebraic signifier as follows:

Ft = At-1 ( 10 )

Where Ft represents the prognosis value for clip period T and At-1 represent the ascertained value on period earlier ( t-1 ) . However, we might reason that in add-on to sing merely the most recent observation, it would do sense to see the way from which we arrived at the latest observation. If the

series is dropped to the latest point, possibly it is sensible to presume some farther bead. Alternatively, if we merely observed an addition, it may do sense to factor into our prognosis some farther addition. This can be adjusted or can be made in the 2nd naA?ve prediction theoretical account, which includes some proportion of the most late observed rate of alteration in the series. In general algebraic footings the theoretical account is as follows:

Ft = At-1 + ( At-1 - At-2 ) ( 11 )

Where Ft is the prognosis for period T, At-1 is the existent observation at period t-1, At-2 is the ascertained value at period t-2.

So, to calculate the demand of ecotourism in Ba'Kelalan for the first one-fourth of 2011 ( denoted as BKM0111 ) , we take the ascertained value for the 4th one-fourth of 2010 ( BKM0410 ) and adjust it by including some information from the most recent tendency. Therefore, the prognosis for BKM0111 harmonizing to the equation ( 11 ) is:

BKM0111 = BKM0410 + ( BKM0410 - BKM0310 )

Table 4: NaA?ve Forecast 2 Method

Year

One-fourth

No. Of Arrival

NaA?ve Trend Forecast

Mistake

Square Error

Absolute Error

% Mistake

% Absolute Error

2010

1

708

263

445

198025

445

0.63

0.63

2

767

1016

-249

62001

249

-0.32

0.32

3

743

826

-83

6889

83

-0.11

0.11

4

532

719

-187

34969

187

-0.35

0.35

2011

321

The computation is as follows:

BKM0111 = 532 + ( 532-743 )

= 532 + ( -211 )

= 321

In this survey, NaA?ve Forecast 1 and NaA?ve Forecast 2 have been obtained to happen the better value of RMSE. As a consequence, it is found that NaA?ve Forecast 1 is better comparison to NaA?ve Forecast 2 as the RMSE is lower which is 123.60 and 196.65 severally.

The mean alteration theoretical account is another method that has applied in this survey which represents the mean of all the informations to calculate. The term traveling norm is used to depict this attack. Basically, three- and five one-fourth traveling mean are normally applied in this attack. As to calculate the demand of ecotourism in Ba'Kelalan, three- and five-moving norm has been done to place which tendency is the best to foretell the flow of tourer to this topographic point in the hereafter. The application is similar to the naA?ve prognosis measure nevertheless we have to cipher the mean of the norm as follows harmonizing to the equation ( 2 ) .

Table 5: Average Change Model

Year

One-fourth

No.Of Arrival

5 Qtr MA

Forecast 5 Qtr MA

Mistake

Square Error

Absolute Error

% Mistake

Absolute % Mistake

2009

1

512

432.4

378

54.4

2959.36

54.4

0.11

0.11

2

701

498.8

432.4

66.4

4408.96

66.4

0.09

0.09

3

537

471.6

498.8

-27.2

739.84

27.2

-0.05

0.05

4

400

510.2

471.6

38.6

1489.96

38.6

0.10

0.10

2010

1

708

571.6

510.2

61.4

3769.96

61.4

0.09

0.09

2

767

622.6

571.6

51

2601

51

0.07

0.07

3

743

631

622.6

8.4

70.56

8.4

0.01

0.01

4

532

630

631

-1

1

1

0.00

0.00

2011

1

630

By utilizing the 5 one-fourth traveling norm, the prognosis for the tourer demand for the one-fourth for 2011 ( BKM0111 ) is:

BKM0111 = 400 + 708 + 767 + 743 + 532 = 3150 = 630

5 5

In this survey, five-quarter moving norm is the best compared to the three-quarter moving mean whereby the RMSE for three-quarter moving norm is 49.52 compared to the five-quarter moving norm which has the smallest RMSE with 29.85.

For the exponential smoothing method, it can be done by utilizing Microsoft Excel every bit good as the naA?ve and mean alteration theoretical account. However, it is easier to analyse it utilizing the Eviews. We can analyse the information utilizing this method by ciphering or calculate it utilizing the equation ( 5 ) . Below is the consequence RMSE of the Exponential Smoothing Method utilizing Eviews:

Table 6: Exponential Smoothing

Method: Simple

Exponential

Parameters:

Alpha

0.478

Sum of Squared Residuals

537351.2

Root Mean Squared Error

110.5103

By utilizing the degree smoothing changeless which is alpha ( I± ) we can continue to calculate utilizing this method. With exponential smoothing, the prognosis value at any clip is a leaden norm of all the available old values. We found that the RMSE of this method is 110.51.

As for the Holt 's method, it can besides be done utilizing the Eviews which consequence is as follows:

Table 7: Holt 's Method

Method: Holt-Winter No Seasonal

Parameters:

Alpha

0.11

Beta

1

Sum of Squared Residuals

418306.8

Root Mean Squared Error

97.50371

From the consequence of analysis, Holt 's method with I± = 0.11 and I? = 1, it ensuing RMSE with 97.50. Advantage by utilizing this method is that the recent observations are given comparatively more weight in calculating the than the older observations.

Decision

Based on the one measure in front forecast analysis, Average Change theoretical account is the most suited method for calculating the quarterly demand of ecotourism. Each theoretical account type has a alone feature which fits to a peculiar information series. More forecasting techniques should be explored to guarantee the fittingness to longer series of demand on ecotourism. Univariate Modelling Technique is fundamentally a individual variable theoretical account that uses past information as the footing to bring forth the prognosis values. This is made on the premise that the prognosis values are dependent entirely on the past form of the information series.