Metode Peramalan Fuzzy Time Series-Markov Chain Berdasarkan K-Means Clustering (Fuzzy Time Series-Markov Chain Forecasting Method Based On K-Means Clustering)

Text

Metode Peramalan Fuzzy Time Series-Markov Chain Berdasarkan K-Means Clustering (Fuzzy Time Series-Markov Chain Forecasting Method Based On K-Means Clustering)

Nega Rananti Retmadika - Nama Orang;

ABSTRAK
METODE PERAMALAN FUZZY TIME SERIES-MARKOV CHAIN
BERDASARKAN K-MEANS CLUSTERING
Oleh
Nega Rananti Retmadika
24010116120029
Sistem peramalan fuzzy time series menangkap pola dari data yang telah lalu,
kemudian digunakan untuk meramalkan data yang akan datang. Dalam
perhitungan peramalan fuzzy time series, panjang interval ditentukan di awal
proses perhitungan, sedangkan penentuan panjang interval sangat berpengaruh
dalam pembentukan fuzzy relationship, sehingga nantinya akan berdampak pada
hasil peramalan. K-means clustering merupakan metode pengelompokan data
yang dapat digunakan untuk menentukan panjang interval berdasarkan nilai pusat
clusternya. Hasil pengelompokan menggunakan k-means dapat diukur tingkat
kekuatan pengelompokannya menggunakan indeks validitas silhouette. Dalam
skripsi ini, fuzzy time series-markov chain berdasarkan k-means clustering
diterapkan untuk meramalkan data harga saham penutupan, pembukaan, dan
tertinggi memberikan hasil yang cukup baik, dengan nilai indeks validitas
silhouette masing-masing data yaitu 0,909, 0,793, dan 0,862, dan nilai AFER
untuk masing-masing data yaitu 1,679%, 1,856%, dan 1,495%.
Kata kunci: peramalan, saham, fuzzy time series, fuzzy time series-markov chain,
k-means clustering, indeks silhouette, AFER.

ABSTRACT
FUZZY TIME SERIES-MARKOV CHAIN FORECASTING METHOD
BASED ON K-MEANS CLUSTERING
by
Nega Rananti Retmadika
24010116120029
Forecasting system with fuzzy time series capturing the pattern from past data,
then use it to predict future data. In the calculastion of fuzzy time series
forecasting, the length of the interval is determined at the beginning of the
calculation process, while determining the interval length is very influential in the
formation of fuzzy relationships, so that later it will have an impact on forecasting
results. K-means clustering is a method of grouping data that can be used to
determine the length of the interval based on the value of the cluster center. The
result of grouping using k-means can be measured the level of strength of the
grouping using the silhouette validity index. In this thesis, fuzzy time seriesmarkov chain based on k-means clustering is applied to predict closing, opening,
and highest stock price data that gives good results, with silhouette validity index
values of each data are 0,909, 0,793, and 0,862, and AFER values for each data is
1,679%, 1,856%, and 1,495%.
Keywords: forecasting, stock price, fuzzy time series, fuzzy time series-markov
chain, k-means clustering, indeks silhouette, AFER.

Ketersediaan

2286A20IV 2286 A 20-iv Perpustakaan FSM Undip (Referensi) Tersedia

Informasi Detail

Judul Seri: MATEMATIKA
No. Panggil: 2286 A 20-iv
Penerbit: : ., 2020
Deskripsi Fisik: -
Bahasa: Indonesia
ISBN/ISSN: -
Klasifikasi: NONE
Tipe Isi: -
Tipe Media: -
Tipe Pembawa: -
Edisi: -
Subjek: -
Info Detail Spesifik: -
Pernyataan Tanggungjawab: Nega Rananti Retmadika

Metode Peramalan Fuzzy Time Series-Markov Chain Berdasarkan K-Means Clustering (Fuzzy Time Series-Markov Chain Forecasting Method Based On K-Means Clustering)

Ketersediaan

Informasi Detail

Versi lain/terkait

Lampiran Berkas

Komentar