International Journal of Statistics and Probability; Vol. 5, No.

6; November 2016
ISSN 1927-7032 E-ISSN 1927-7040
Published by Canadian Center of Science and Education

Seasonal Modelling of Fourier Series with Linear Trend

Iberedem A. Iwok1
1
Department of Mathematics/Statistics, University of Port-Harcourt, P.M.B.5323, Port-Harcourt, Rivers State, Nigeria.
Correspondence: Iberedem A. Iwok, Department of Mathematics/Statistics, University of Port-Harcourt, P.M.B.5323,
Port-Harcourt, Rivers State, Nigeria. E-mail: ibywok@yahoo.com

Received: July 28, 2016 Accepted: September 6, 2016 Online Published: October 26, 2016
doi:10.5539/ijsp.v5n6p65 URL: http://dx.doi.org/10.5539/ijsp.v5n6p65

Abstract
This work was motivated by the need to model a periodic time series function with linear trend. A Fourier series
representation with detrended linear function was proposed. In this representation, the time series 𝑋𝑡 is expressed as a
combination of the linear trend component and a linear combination of 𝑠 orthogonal trigonometric functions; where 𝑠 is
the number of seasons. The method was applied to a rainfall data and the proposed model was found to give a good fit.
Comparative study was carried out with the complete Fourier representation. Diagnostic checks revealed that the
proposed method performs better the pure Fourier approach.
Keywords: Fourier series, seasonal model, linear trend, periodogram, spectral density and white noise process.

1. Introduction
The usual autoregressive integrated moving average (ARIMA) models developed by Box and Jenkins (1970) has been
extensively used in modelling linear time series. The ARIMA models assume that the current observation depends on
weighted previous observations, weighted previous random shocks and the current shock. However, most time series
arising in nature do not assume linearity but rather, periodic or seasonal with linear trend. Seasonal time series contain a
seasonal phenomenon that repeats itself after a regular period of time. Such phenomena stem from factors such as weather,
which affects many business and economic activities, cultural events and graduation ceremonies. Series with seasonal
pattern cannot be adequately represented by ARIMA models. To analyze such series, Wold (1974) arranged the series in a
two dimensional table according to the season; and the totals and averages were computed. In Wold (1974) representation,
a time series is thought to consist of trend-cycle, seasonal and irregular components. To estimate these components,
several decompositions are usually involved. Box, Jenkins and Reinsel (2008) made an extension of the Box and Jenkins
(1970) ARIMA models to include the seasonal part and is called the seasonal autoregressive integrated moving average
(SARIMA) models. Despite these efforts, the models do not adequately represent most periodic series.
A better procedure extensively used for modelling periodic time series is the Fourier analysis. This method represent the
time series by a set of elementary functions called basis such that all functions under study can be written as linear
combinations of the elementary functions in the basis. These elementary functions involve the sine and cosine functions
or complex exponentials. The Fourier series approach describes the fluctuation of time series in terms of sinusoidal
behaviour at various frequencies. Despite the wider acceptability of the method, however, Fourier approach still suffers
some set backs. One major problem associated with it is the cumbersomeness in Fourier representation and non inclusion
of trend component. As will be seen in the methodology, the inconveniences in representing the time series is enormous if
we are to include all the terms required in Fourier series. This cannot go well with a series of large sample size because
representing all the terms will consume several pages and can be boring to both the researcher and the reader. Hence, there
is need to shorten the number of terms in the Fourier expression and give a summarized representation that adequately
describe the time series. This is the intent of this work. As earlier stated, seasonal variations in time series can be caused
by climatic factor and we are going to use rainfall data in our illustration.
2. Literature Review
The need for accurate rainfall prediction is necessary when considering the importance in which such information would
give for river control, reservoir operation, forestry interests, flood mitigation, etc. Due to the numerous benefits of rainfall
modelling and prediction, studies on rainfall analysis have been on the fore front in the research world.
Afshar, Joshua, Buckman, and Samuel (2014) modelled rainfall data using ARIMA model and artificial neural networks
(ANNs). The ARIMA was found to give discouraging result. However, with application of artificial neural network to the

65
http://ijsp.ccsenet.org International Journal of Statistics and Probability Vol. 5, No. 6; 2016

ARIMA component, the combined model was found to be adequate and forecast were generated.
Cenis (1989) studied temperature in solarized soil using Fourier analysis. He obtained the daily maximum and minimum
temperatures at two dept on daily basis for three summer periods. He used the values to fit sinusoidal equations which
accounted for 93% variation. The variation and the hourly mean differences between the measured temperatures were
calculated. The analysis gave an overall encouraging result.
Serangelo, Ferrari and De Luca (2011) applied non-homogeneous Poisson process to examine the seasonal effects of daily
rainfall. The modelling process involved the partitioning of observed daily rainfall data into calibration periods for the
estimation of parameters. Though the validation period for checking the occurrence process changed; the model which
was applied to the set of rain gauges placed at different geographical areas was shown to provide good fit.
Falahah and Suorapto (2010) carried out research on rainfall data using analytic factor method. The data was obtained
from 50 weather stations for a period of 30 years. The result was plotted on pattern factors to reveal dominant factor for
each region and inspection period. The method explained factors that influence rainfall in Indonesia and the reasons for
having relatively high humidity in one area than the other.
Necholas, Mahmood and Hazan (2013) modelled rainfall data amounts for agriculture planning using gamma distribution
models. Daily rainfall data of two stations having two different mean annual rainfalls were analyzed. Generalized linear
models were used to fit smooth regression curves. The mean amount of rain per rainy day was computed using the
estimates of parameters of the model for each day of the season. The adequacy of the fitted model was check by the
analysis of deviance residuals and was found to be satisfactory. Fourier approach was employed for comparative study. It
was discovered that though reasonable results were obtained, Fourier analysis was time consuming and boring. However,
Fourier series was found suitable in fitting gamma distribution for the determination of mean rain per rainy day.
Zakaria (2013) conducted a study on periodic and stochastic modelling of monthly rainfall and the periodicities were
determined. Stochastic components were estimated using the auto-regressive model approach. Residuals obtained from
the model were shown to follow a white noise process; thus indicating the adequacy of the fitted model.
Beatrice, Nasser, Afshar, Selaman and Fahmi (2014) analyzed data from eight rain gauge stations. Annual rainfall data for
27 years were computed with the Fourier series equation. The result was compared with that obtained from harmonic
series models. It was discovered that both models were capable of describing rainfall pattern and were able to provide
reasonable relationship between the simulated and the observed data.
Akpanta, Okorie and Okoye (2015) adopted SARIMA modelling of the frequency approach in analyzing monthly rainfall
data in Umuahia. Probability time series approach was considered. The original data plotted showed seasonality which
was removed by differencing. After subjecting the model to diagnostic checks, SARIMA (0,0,0)(0,1,1)12 was found to fit
the data well and was used for prediction.
3. Methodology
In this method, a periodic time series is first observed whether it contains a linear trend or not. Visual inspection of the raw
data plot can reveal this pattern. Assuming a linear trend is detected, a linear regression model of the form
𝑌𝑡 = 𝛽0 + 𝛽1 𝑡 + 𝑒𝑡 (1)
is first fitted to the data;
where 𝑌𝑡 is the observed time series, 𝑡 is the time points (𝑡 = 1,2, … , 𝑛), 𝑛 is the number of observations, 𝛽0 and 𝛽1
are the regression parameters, and 𝑒𝑡 is the error component.
Fitting the above model (1) to the data 𝑌𝑡 , we can obtain the estimate of the error component
̂0 − 𝛽
𝑒̂𝑡 = 𝑌̂𝑡 − 𝛽 ̂1 𝑡
which can be tested for randomness or white noise.
̂0 + 𝛽
After obtaining the trend equation ( i.e. 𝑌̂ = 𝛽 ̂1 𝑡 ), the main series 𝑌𝑡 is detrended by the expression
𝑦𝑡 = 𝑌𝑡 − 𝑌̂𝑡 = 𝑌𝑡 − 𝛽̂0 − 𝛽̂1 𝑡 (2)
The resulting series 𝑦𝑡 is then used to fit seasonal model using Fourier representation.
3.1 Fourier Series Representation of the Time Series 𝒚𝒕
Given a time series of 𝑛 observations, the Fourier representation is the set of 𝑞 orthogonal trigonometric functions
shown below:
𝑞
𝑦𝑡 = ∑𝑖=1(𝛼𝑖 cos2𝜋𝑓𝑖 𝑡 + 𝛽𝑖 sin2𝜋𝑓𝑖 𝑡 ) + 𝑒𝑡 (3)

66
http://ijsp.ccsenet.org International Journal of Statistics and Probability Vol. 5, No. 6; 2016

estimated by
𝑞
𝑦̂𝑡 = ∑𝑖=1(𝑎𝑖 cos2𝜋𝑓𝑖 𝑡 + 𝑏𝑖 sin2𝜋𝑓𝑖 𝑡 ) (4)
2 2
where 𝑞 = 𝑛⁄2, 𝑎𝑖 = ∑𝑛𝑡=1 𝑦𝑡 cos2𝜋𝑓𝑖 𝑡 , 𝑏𝑖 = ∑𝑛𝑡=1 𝑦𝑡 sin2𝜋𝑓𝑖 𝑡,
𝑛 𝑛

𝑒𝑡 ~𝑁𝐼𝐼𝐷(0, 𝜍 2 ); period = 𝑝𝑖 = 𝑛⁄𝑖 and 𝑓𝑖 = 𝑖⁄𝑛 is the 𝑖 𝑡ℎ harmonic of the fundamental frequency 1⁄𝑛.
3.2 The Peridogram
The periodogram is defined as the function of intensities 𝐼(𝑓𝑖 ) at frequency 𝑓𝑖 = 𝑖/𝑛 and is given as
𝑛
𝐼(𝑓𝑖 ) = (𝑎𝑖2 + 𝑏𝑖2 ) ; 𝑖 = 1,2, … , 𝑞.
2

Periodogram is the plot of the intensities against the frequencies or periods. The periodogram 𝐼(𝑓𝑖 ) is simply the sum of
squares associated with the pair of coefficients (𝑎𝑖 , 𝑏𝑖 ) and hence with the frequency 𝑓𝑖 or period 𝑝𝑖 . That is,
∑𝑞𝑡=1(𝑦𝑡 − 𝑦)2 = ∑𝑛/2
𝑡=1 𝐼(𝑓𝑖 ).
In the context at hand, the periodogram is used to determine the seasonality or periodicity of a time series. This is usually
indicated by the largest peak in the periodogram plot.
3.3 The Spectrum
The sample spectrum is obtained by allowing the frequency 𝑓 to vary continuously in the range 0 to 0.5 cycle so that the
periodogram can be re-defined as
𝑛
𝐼(𝑓) = (𝑎𝑗2 + 𝑏𝑗2 ) ; 0 ≤ 𝑓 ≤ 0.5.
2

The function 𝐼(𝑓) is called the spectrum.

3.4 Autocorrelation Function
This is the plot of autocorrelation at lag k (𝜌𝑘 ) versus k.
3.5 Spectral Density Function
Spectral density is the Fourier transform of the auto-correlation function and is estimated by
𝑔(𝑓) = 2,1 + ∑∞ 𝑘=1 𝜌𝑘 𝑐𝑜𝑠(2𝜋𝑓𝑘)- ; 0 ≤ 𝑓 ≤ 0.5
where 𝜌𝑘 is the autocorrelation at lag k. The spectral density performs the same function as the periodogram. The period
or seasonality of a time series is obtained at where the spectral density is maximum.
3.6 White Noise Process
A process *𝜀𝑡 + is said to be a white noise process with mean 0 and variance 𝜍𝜀2 written *𝜀𝑡 +~𝑊𝑁(0, 𝜍𝜀2 ), if it is a
sequence of uncorrelated random variables from a fixed distribution.
3.7 The Seasonal Fourier Representation
Rather than fitting the entire Fourier series expression in equation (3), we fit only the Fourier terms up to the season
detected by the periodogram. That is, suppose the season determined by the periodogram in the detrended series 𝑦𝑡 is 𝑠,
then equation (3) reduces to
𝑦𝑡 = ∑𝑠𝑖=1(𝛼𝑖 cos2𝜋𝑓𝑖 𝑡 + 𝛽𝑖 sin 2𝜋𝑓𝑖 𝑡) + 𝜀𝑡 (5)
̂0 + 𝛽
⟹ 𝑌̂𝑡 = 𝛽 ̂1 𝑡 + ∑𝑠𝑖=1(𝑎𝑖 cos2𝜋𝑓𝑖 𝑡 + 𝑏𝑖 sin 2𝜋𝑓𝑖 𝑡) (6)
and
𝜀̂𝑡 = 𝑌𝑡 − 𝑌̂𝑡 (7)
Comparatively, the expression (5) is less cumbersome in carrying out analysis than the complete Fourier form expressed
in equation (3). The model (5) can be fitted to any periodic or seasonal data and the estimated residuals 𝜀̂𝑡 obtained from
(7) can be tested for white noise to determine whether the model is adequate or not.
4. Data Analysis and Result
The data used for this work is the average monthly rainfall data (𝑌𝑡 ) in Calabar, Nigeria between 2005-2015 (Source:
www.cbn.gov.ng); and the analysis is carried out using Minitab and gretl softeware.

67
http://ijsp.ccsenet.org International Journal of Statistics and Probability Vol. 5, No. 6; 2016

4.1 Complete Fourier Series Model

600

120
Fitting the full Fourier
500 series in equation (3) where 𝑞 = = 60 result in a residual variance of 11.23 and the Fourier
2

400
coefficients are displayed in Appendix C. The residual autocorrelation function is displayed in figure 5. Clearly, there is a
at lag 12 ( 𝜌𝑘 = −0.36). This shows that the residuals are correlated (at lag 12) and hence do not follow
Yt

significant spike300
a white noise process. The actual and estimate values plots displayed in figure 6 shows a low correlation between these
200
values. Thus, the full Fourier series, despite it cumbersome nature does not fit adequately to the data.
100
4.2 The Proposed Approach
4.2.1 Seasonality and
0 the Estimated Trend
Index 20
The raw data plot in figure 1 clearly shows the40existence 60 80
of seasonality 100
and trend. 120
This is indicated by the periodic pattern
and upward movement of the graph. Fitting the trend equation gives the Minitab output in table 1 below.

600

500

400
Yt

300

200

100

Index 20 40 60 80 100 120

Figure 1. Raw data plot of the series 𝑌𝑡

Table 1. Minitab Output for the Trend equation
Predictor Coef StDev T P
Constant 185.28 26.96 6.87 0.000
t 0.6639 0.3898 1.70 0.031
The regression equation is Yt = 185 + 0.6639 t
The p–values in table 1 shows that both the constant term and the coefficient of 𝑡 are significant. Thus, the estimated
trend equation is
̂
𝑌𝑡 = 185.28 + 0.6639𝑡
Next, we obtain the new series from equation (2) as
𝑦𝑡 = 𝑌𝑡 − 𝑌̂𝑡 = 𝑌𝑡 − 185.28 − 0.6639𝑡
The new series 𝑦𝑡 now becomes our working data.
4.2.2 The Peridogram and Spectral Density of 𝒚𝒕
The periodogram analysis was conducted using the gretl software. The periodogram is displayed in figure 2. The values
for the spectral densities, periods and frequencies are equally displayed in appendix A. From the table of appendix A and
figure 2; it is observed that the periodogram is dominated by a very large peak at scaled frequency, 𝑓𝑖∗ = 𝑖 = 10
(𝑓𝑖 = 𝑖/𝑛 = 10/120 = 0.0833). This is indicated by the largest spectral density of 16.397 shown in bold figures. This
density and frequency correspond to a period or season, 𝑠 = 𝑛/𝑖 = 120/10 = 12 months. This indicates a 12 – month
cycle.

68
http://ijsp.ccsenet.org International Journal of Statistics and Probability Vol. 5, No. 6; 2016

Spectrum of yt

periods
120.0 10.9 5.7 3.9 2.9 2.4

18

16

14

12

10

0
0 10 20 30 40 50 60
scaled frequency

Figure 2. Periodogram plot of 𝑦𝑡

4.2.3 The Seasonal Fourier Representation of 𝒚𝒕
Rather than use 𝑞 = 60 in equation (3), we reduce the Fourier components to the number of seasons, 𝑠 = 𝑞 = 12 and
apply equation (5). That is,
𝑦𝑡 = ∑12
𝑖=1(𝛼𝑖 cos2𝜋𝑓𝑖 𝑡 + 𝛽𝑖 sin 2𝜋𝑓𝑖 𝑡) + 𝜀𝑡 (8)
Equivalently, equation (8) can be expressed as
𝑦𝑡 = ∑12
𝑖=1(𝛼𝑖 cos𝜔𝑖 𝑡 + 𝛽𝑖 sin 𝜔𝑖 𝑡) + 𝜀𝑡 (9)
where, 𝜔𝑖 = 2𝜋𝑓𝑖 .
Subjecting equation (9) to regression analysis give the parameter estimates displayed in Appendix B. In appendix B, some
coefficients of the variables are not statistically significant (i.e. their 𝑝 values are less than 0.05) and are therefore
excluded in the overall equation.
Hence, from (6), the resulting estimated model is
𝑌̂𝑡 = 185.28 + 0.6639𝑡 − 0.621 sin 𝜔1 𝑡 − 0.314 sin 𝜔2 𝑡 − 0.129 sin 𝜔3 𝑡 − 0.105 sin 𝜔4 𝑡
−0.085 sin 𝜔5 𝑡 − 0.081 cos 𝜔8 𝑡 − 0.155 cos 𝜔9 𝑡 − 0.074 sin 𝜔10 𝑡
+ 0.142 cos 𝜔11 𝑡 − 0.206 cos 𝜔11 𝑡 + 0.094 cos 𝜔12 𝑡 (10)

The residual 𝜀̂𝑡 of the fitted is obtained from

𝜀̂𝑡 = 𝑌𝑡 − 𝑌̂𝑡
5. Diagnosis
We present here two diagnostic checks to ensure that the proposed model (10) fitted to the data is adequate.
5.1 Actual and Estimate Plots
The overlaid plots of the actual values (𝑌𝑡 ) and the estimated values (𝑌̂𝑡 ) is displayed in figure 3. The two superimposed
plots move together in the same direction indicating closeness and strong correlation between the values of the two
variables. This shows that the model is adequate.

69
http://ijsp.ccsenet.org International Journal of Statistics and Probability Vol. 5, No. 6; 2016

Time Series Plot of Yt and Fit(Yt)

600

500

400
Yt

300

200

100

Index 20 40 60 80 100 120

Figure 3. Actual and estimate plot of 𝑌𝑡

5.2 Residual Variance, Autocorrelation and White Noise
The residual variance of the fitted model is 4.78. This is significantly smaller than the 11.23 obtained from fitting the full
Fourier series. The residual autocorrelation function is displayed in figure 4 and it shows that there is no significant
autocorrelation of the residuals. That is, all autocorrelations at all lags are within the range ±2/√𝑛 (as indicated by the
two red lines in figure 4). This means the residuals of the fitted model are not serially correlated. In more precise terms,
the residuals follow a white noise process. Hence, the model is adequate.

Autocorrelation Function for Res(yt)

1.0
0.8
Autocorrelation

0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

10 20 30

Lag Corr T LBQ Lag Corr T LBQ Lag Corr T LBQ Lag Corr T LBQ

1 -0.01 -0.08 0.01 10 -0.05 -0.47 8.44 19 0.03 0.29 20.11 28 -0.09 -0.87 27.79
2 0.14 1.52 2.41 11 0.11 1.11 9.99 20 -0.07 -0.65 20.79 29 0.02 0.15 27.83
3 0.09 0.96 3.41 12 -0.14 -1.39 12.51 21 -0.06 -0.54 21.27 30 0.01 0.09 27.85
4 0.13 1.36 5.46 13 0.13 1.30 14.81 22 -0.08 -0.73 22.13
5 0.05 0.54 5.80 14 0.01 0.11 14.83 23 0.01 0.07 22.14
6 -0.09 -0.92 6.79 15 0.13 1.30 17.25 24 -0.11 -1.06 24.04
7 -0.09 -0.93 7.83 16 -0.01 -0.08 17.26 25 -0.11 -1.07 26.02
8 0.03 0.28 7.93 17 -0.04 -0.40 17.50 26 -0.02 -0.18 26.08
9 -0.04 -0.43 8.16 18 0.13 1.28 19.97 27 0.05 0.43 26.41

Figure 4. Residual Autocorrelation function plot of 𝜀̂𝑡

6. Discussion and Conclusion
It has been noted that the Fourier series model can only be applied to periodic series that are stationary in mean. If the
series contain trend, however, special technique is required for the modelling process. Perhaps, this constituted the
problem of Afshar et al (2014) that made them to obtain a discouraging result by applying ARIMA model to a periodic
data without considering the trend. Besides, as noted by Necholas et al (2013), Fourier series modelling is time consuming
and boring because of the large number of Fourier coefficients involved. In this work, however, these problems have been
addressed. As clearly demonstrated in this method, the trend component of a periodic series is taken care of. Also, setting
𝑞 = 𝑠 has reduced the computational burden by 80% and has given adequate Fourier representation as confirmed by the
diagnostic checks. It is believed that this work has opened another possibility of addressing periodic functions.

70
http://ijsp.ccsenet.org International Journal of Statistics and Probability Vol. 5, No. 6; 2016

Autocorrelation Function for RES(Yt*)

1.0
Autocorrelation 0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

10 20 30

Lag Corr T LBQ Lag Corr T LBQ Lag Corr T LBQ Lag Corr T LBQ

1 -0.09 -1.00 1.02 10 0.00 0.01 9.71 19 0.02 0.21 36.28 28 -0.07 -0.60 40.84
2 0.12 1.30 2.79 11 0.11 1.16 11.46 20 -0.09 -0.79 37.45 29 -0.02 -0.20 40.92
3 0.04 0.48 3.04 12 -0.36 -3.67 29.42 21 -0.02 -0.18 37.52 30 -0.08 -0.67 41.89
4 0.06 0.67 3.53 13 0.17 1.51 33.18 22 -0.09 -0.78 38.70
5 0.05 0.56 3.89 14 0.03 0.25 33.29 23 0.06 0.52 39.25
6 -0.15 -1.62 6.88 15 0.03 0.29 33.44 24 0.02 0.15 39.29
7 -0.14 -1.50 9.56 16 -0.02 -0.15 33.48 25 -0.07 -0.60 40.02
8 -0.01 -0.15 9.59 17 -0.09 -0.82 34.66 26 -0.00 -0.00 40.02
9 -0.03 -0.31 9.71 18 0.10 0.92 36.20 27 0.02 0.17 40.08

Figure 5. Residual Autocorrelation function plot of 𝑒̂𝑡

Time Series Plot of Yt and Fit(Yt*)

700

600

500

400
Yt

300

200

100

Index 20 40 60 80 100 120

Figure 6. Actual and estimate plot of 𝑌𝑡∗

References
Afshar, B. N., Joshua, M. A, Buckman, A., & Samuel, T. (2014). Time series modeling of rainfall in new Juaben
municipality of the Eastern region of Ghana. Special issue on contemporoary research in business and social science,
4(7), 21-28.
Akpanta, C. A., Okorie, I. E., & Okoye, N. N. (2015). SARIMA modelling of the frequency of monthly rainfall in
Umuahia, Abia state of Nigeria. American journal of mathematics and statistics, 5, 82-87.
Beatrice, C. B., Nasser, R., Afshar, K., Selaman, O., & Fahmi, H. (2014). Application of mathematical modelling in
rainfall forecast: A case study in SGS. Sarawak basin. International Journal of Research in Engineering and
Technology, 3, 2321-7308.
Box, G. E. P., & Jenkins, G. M. (1970). Time Series Analysis; forecasting and control. Holden Day. San Francisco,
California.
Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (2008). Time Series Analysis; Forecasting and Control. John Willey &
Sons.
Cenis, J. L. (1989). Temperature evaluation in solarized soils by Fourier analysis. Photopathology, 79, 506-510.
http://dx.doi.org/10.1094/Phyto-79-506
Falahah, M., & Suprapto, S. (2010). Interpretation of rainfall data using analysis factor method. Proceeding of third

71
http://ijsp.ccsenet.org International Journal of Statistics and Probability Vol. 5, No. 6; 2016

international conference on mathematics and natural sciences, Indonesia (pp. 1288-1293).

Necholas, Z., Mahmood, Z., & Hazan, Y. (2013). Modeling the daily rainfall amounts of north-west Pakistan for
agricultural planning. Sarhad J. Agric, 30, 5-23.
Serangelo, B. A., Ferrari, H. K., & DeLuca, T. (2011), Occurrence analysis of daily rainfall through non-homogeneous
Poisson process. Nat. Hazards Earth Syst. Sci., 11,1657-2011. http://dx.doi.org/10.5194/nhess-11-1657-2011
Wold, H. O. A. (1974). A study in the analysis of stationary time series. Almqvist and Wiksell, Uppsala.
Zakaria, A. (2013). A study of periodic and stochastic modelling of monthly rainfall from Purajaya station. Asian
Transactions on Engineering, 1(3), 1-7.

Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/4.0/).

72