Implementasi High Order Intuitionistic Fuzzy Time Series Pada Peramalan Indeks Harga Saham Gabungan

Penulis

  • Titis Jati Nugraha Universitas Sebelas Maret, Surakarta
  • Winita Sulandari Universitas Sebelas Maret, Surakarta
  • Isnandar Slamet Universitas Sebelas Maret, Surakarta
  • Sri Subanti Universitas Sebelas Maret, Surakarta
  • Etik Zukhronah Universitas Sebelas Maret, Surakarta
  • Sugianto Sugianto Universitas Sebelas Maret, Surakarta
  • Irwan Susanto Universitas Sebelas Maret, Surakarta

DOI:

https://doi.org/10.25126/jtiik.20241127363

Abstrak

Indeks Harga Saham Gabungan (IHSG) adalah indeks yang mengukur kinerja harga semua saham yang terdaftar di Bursa Efek Indonesia (BEI). Pergerakan IHSG menjadi acuan para investor untuk menetapkan keputusan finansial yang berkaitan dengan untung rugi dalam berinvestasi. Oleh karenanya, informasi peramalan IHSG yang akurat sangat penting bagi para investor. Penelitian ini membahas penerapan metode High Order Intuitionistic Fuzzy Time Series (HOIFTS) dalam peramalan IHSG di BEI. Metode HOIFTS melibatkan tiga indikator, yaitu derajat keanggotaan, derajat non-keanggotaan, dan fungsi skor (indeks intutionistic) sehingga model yang dihasilkan mampu menangani ketidakpastian dalam data. Tahapan penting dalam pemodelan HOIFTS adalah pada intuitionistic fuzzification, penentuan relasi logika fuzzy intutionistic, dan proses intutionistic defuzzification order tinggi. Penelitian ini menetapkan metode Chen, baik order satu maupun order tinggi sebagai metode pembanding untuk melihat seberapa jauh keberhasilan metode HOIFTS dalam meramalkan data bulanan IHSG. Perbandingan nilai RMSE (root mean square error) dan MAPE (mean absolute percentage error) yang dihasilkan oleh model HOIFTS dan dua model benchmark, yaitu Chen order satu dan Chen order tinggi, menunjukkan bahwa metode HOIFTS memiliki nilai kesalahan yang paling kecil yakni nilai RMSE adalah sebesar 57,042 dan MAPE sebesar 0,837% pada data training, sedangkan pada data testing diperoleh nilai RMSE sebesar 38,466 dan MAPE sebesar 0,487%. Dengan demikian, metode HOIFTS lebih direkomendasikan dalam peramalan IHSG dibandingkan dua metode lain yang dibahas dalam penelitian ini.

 

Abstract

The Composite Stock Price Index (CSPI) is an index that measures the price performance of all shares listed on the Indonesia Stock Exchange (ISE). CSPI is a reference for investors to determine financial decisions related to profit and loss in investing. Therefore, accurate CSPI forecasting information is very important for investors.  This research discusses the application of the HOIFTS method in forecasting CSPI on the ISE. The HOIFTS method involves three indicators, namely degree of membership, degree of non-membership, and a score function (intuitionistic index) so that the resulting model is able to handle uncertainty in the data. Important stages in HOIFTS modeling are intuitionistic fuzzification, determination of intuitionistic fuzzy logic relations, and the intuitionistic higher order defuzzification process. This research determines the Chen method, both first order and high order as a comparison method to see how successful the HOIFTS method is in predicting monthly CSPI data. The comparison results of the RMSE (root mean square error) and MAPE (mean absolute percentage error) values ​​produced by the HOIFTS and two benchmark models, i.e., the first order Chen’s and high-order Chen’s, show that the HOIFTS method yields the smallest error value, namely the RMSE value is 57.042 and the MAPE is 0.837% on the training data, whereas in testing data obtained an RMSE value of 38.466 and a MAPE of 0.487%. Thus, the HOIFTS method is more recommended in forecasting CSPI compared to the other two methods discussed in this research.

 

 

Downloads

Download data is not yet available.

Referensi

ABHISHEKH, GAUTAM, S.S. & SINGH, S.R., 2018. A refined method of forecasting based on high-order intuitionistic fuzzy time series data. Progress in Artificial Intelligent, 7(4), pp.339–350.

ABHISHEKH, GAUTAM, S.S. & SINGH, S.R., 2020. A new method of time series forecasting using intuitionistic fuzzy set based on average-length. Journal of Industrial and Production Engineering, 37(4), pp. 175-185.

ALAM, N.M.F.H.N.B., RAMLI, N., & MOHAMAD, D. 2021. Fuzzy time series forecasting model based on intuitionistic fuzzy sets and arithmetic rules. AIP conference proceedings, 2365(1), 050003.

AMRY, Z., & SIREGAR, B.H., 2019. ARIMA model selection for composite stock price index in Indonesia stock exchange. International Journal of Accounting and Financial Studies, 2(1), pp. 31-38

ASHRAF, S., SOHAIL, M., CHOHAN, M.S., PAOKANTA, S., & PARK, C. 2024. Higher-order circular intuitionistic fuzzy time series forecasting methodology: Application of stock change index. Demonstratio Mathematica, 57(1), p.20230115.

CHEN, S.M., 1996. Forecasting enrollments based on fuzzy time series. Fuzzy sets and systems, 81(3), pp.311-319.

CHEN, S.M., 2002. Forecasting enrollments based on high-order fuzzy time series. Cybernetics and Systems, 33(1), pp.1-16.

CHRISANTAMA, A.R., SULANDARI, W., & SUGIYANTO. 2021. Penerapan metode auto singular spectrum analysis pada peramalan data indeks harga gabungan di Indonesia. Prosiding Konferensi Nasional Matematika XX, pp. 405-410.

FITRIA, A. 2021. Peramalan indeks harga saham gabungan menggunakan metode fuzzy time series Markov chain selama pandemi covid-19. Disertasi Doktor, Universitas Jenderal Sudirman.

GOLYANDINA, N., & Zhigljavsky, A., 2020. Singular spectrum analysis for time series. Springer Brief in Statistics. Berlin: Springer.

HANKE, J.E. & WICHERN, D., 2014. Business Forecasting. 9th ed. Edinbrugh Gate: Pearson Education Limited.

ILAFI, A.K., JOWANTI, L., & FADHILAH, A.N. 2020. Pemanfaatan bid data dalam memprediksi harga saham di era new normal. Prosiding seminar nasional Official Statistics, 2020(1), pp. 281-291.

INVESTING.COM, 2022. Data Historis Jakarta Stock Exchange Composite. Fusion Media Limited. < https://id.investing.com/indices/idx-composite-historical-data> [Diakses 15 Mei 2022].

MIFTAHUDDIN, ADILA N.H., DILLA, P.P., INDRIYANI, C.F., & GUNAWAN, E. 2023. Forecasting the composite stock price index in pandemic Covid-19 in Indonesia using ARDL(p,q) model. AIP Conf. Proc. 2975(1), 080003.

PRAMESTI, A.R., SULANDARI, W., SUBANTI, S., & YUDHANTO, Y. 2023. Peramalan Indeks Harga Saham Gabungan dengan menggunakan Metode Fuzzy Time Series Tipe 2. RADIANT: Journal of Applied, Social, and Education Studies, 4(2), pp. 118-133.

ROSALYN, A.M. 2018. The effect of rupiah exchange rate and inflation rate towards composite stock price index in Indonesia stock exchange. Russian Journal of Agricultural and Socio-Economic Sciences. 78(6), pp.53-58,

SONG, Q. & CHISSOM, B., 1994. Forecasting Enrollments with Fuzzy Time Series part 1. Fuzzy Sets and System, 54, pp.1-9.

SULANDARI, W., SUHARTONO, dan YUDHANTO, Y. 2020. Aplikasi Fuzzy Pada Pemodelan Runtun Waktu. Bandung: Khazanah Intelektual.

XIHAO, S. & YIMIN, L., 2008. Average-Based Fuzzy Time Series Models for Forecasting Shanghai Compound Index. ISSN 1 746-7233, England, UK. World Journal of Modelling and Simulation, 4(2), pp.104-111.

XU, Z.S. & YAGER, R.R., 2006. Some geometric aggregation operators based on intuitionistic fuzzy sets. Int. J. Gen. Syst, 35, pp.417– 433.

YOLCU, O.C., EGRIOGLU, E., BAS, E. dan YOLCU, U. 2022. Multivariate intuitionistic fuzzy inference system for stock market prediction: The cases of Istanbul and Taiwan. Applied Soft Computing, 116, p.108363.

WAHYUDI, S.T., 2017. The ARIMA model for the Indonesia Stock Price. International Journal of Economics and Management, 11(S1), pp. 223-236.

Unduhan

Diterbitkan

26-08-2024

Terbitan

Bagian

Ilmu Komputer

Cara Mengutip

Implementasi High Order Intuitionistic Fuzzy Time Series Pada Peramalan Indeks Harga Saham Gabungan. (2024). Jurnal Teknologi Informasi Dan Ilmu Komputer, 11(2), 257-264. https://doi.org/10.25126/jtiik.20241127363