IDENTIFIKASI VARIABEL YANG MEMPENGARUHI BESAR PINJAMAN DENGAN METODE POHON REGRESI (Studi Kasus di Unit Pengelola Kegiatan PNPM Mandiri)

LUQMAN, SHAUMAL (2015) IDENTIFIKASI VARIABEL YANG MEMPENGARUHI BESAR PINJAMAN DENGAN METODE POHON REGRESI (Studi Kasus di Unit Pengelola Kegiatan PNPM Mandiri). Undergraduate thesis, FSM Universitas Diponegoro.

[img]
Preview
PDF
864Kb

Abstract

Most people need a loan to fullfil their daily needs, such as a loan of goods or money. Loan can also be obtained from financial institutions or individuals. In order to the loan granted by a financial institutions is not wrong target, financial institutions usually apply precaution principle. In making decisions related to how much a decent loan granted to a customer, the financial institutions often use the help of statistical methods. One methods often used is the Classification and Regression Trees (CART). Classification and Regression Trees (CART) is a nonparametric method that can be used to identify the variable that affect the amount of the loan at a financial institutions and estimate how much worth of loans granted. Because of the loan is a continous variable so the form of the tree is a Regression Tree. In this thesis, the financial institutions is UPK PNPM Mandiri Mekar Sejati in Kecamatan Bawang Kabupaten Batang. Variables that may be affected for large loans are age, occupation, type of warranty, the number family members, and the average income per month. The analysis showed that the variables that most influence on the income of the loans. MAPE (Mean Absolute Percentage Error) value from this method is 36%. Keyword : Regression tree, CART, Large loans.

Item Type:Thesis (Undergraduate)
Subjects:H Social Sciences > HA Statistics
Divisions:Faculty of Science and Mathematics > Department of Statistics
ID Code:47467
Deposited By:Mr Hasbi Yasin
Deposited On:01 Feb 2016 15:05
Last Modified:01 Feb 2016 15:05

Repository Staff Only: item control page