Hasil Pencarian  ::  Simpan CSV :: Kembali

Abstrak :
The feed forward neural network (FFANN) model has been the most popular form of artificial neural network model used for forecasting, particularly in economics and finance. In this paper, we elucidate the application of FFANN as a means of modeling financial data. We particularly focus on the model building of FFANN as time series model and use inflation rates in Indonesia as a case study. A comparison is drawn between FFANN model and the best existing models based on traditional econometrics time series approach. The best models are selected on forecasting ability by using the MSE, particularly on the dynamic forecast. The results show that FFANN models outperform the traditional econometric time series model.

Abstrak :
ABSTRAK

Penelitian ini dibuat dalam rangka meneliti perbaikan sistem pengawasan wajib pajak yang
didasarkan dari adanya potensi ketidakpatuhan wajib pajak yang dilihat dari nilai pengungkapan
harta tax amnesty sebesar Rp4.884 Triliun dan rendahnya tax ratio yang pada tahun 2017 sebesar
10.7%. Sistem self assessment, Penyidikan, Pemeriksaan, dan Pengawasan oleh Account
Representative masih menunjukkan kelemahan dengan tidak terdeteksinya nilai dari deklarasi harta
tersebut. Penelitian ini menggunakan theory planned behavior untuk menjelaskan adanya
ketidakpatuhan wajib pajak. Dalam mengontrol ketidakpatuhan tersebut Direktorat Jenderal Pajak
dapat menggunakan data informasi yang akan dibandingkan dengan SPT Wajib Pajak yang datanya
dapat diperoleh berdasarkan Undang-Undang Nomor 9 Tahun 2017 dan Peraturan Pemerintah
Nomor 31 Tahun 2012. Penelitian ini disusun dengan metode deskriptif dengan melakukan
wawancara, perbandingan sistem yang ada dengan yang akan diajukan, dan penelitian terhadap
peraturan perundangan yang berlaku. Penelitian ini menunjukkan perlunya pengawasan yang lebih
detail lagi kepada wajib pajak, sehingga ketidakpatuhan wajib pajak dalam melakukan pembayaran
dan pelaporan pajak dapat diminimalisir.

---

ABSTRACT

---

This study was made in order to examine the improvement of the taxpayer supervision system
based on the potential non-compliance of taxpayers as seen from the value of tax amnesty
disclosure of Rp.4,884 Trillion and the low tax ratio in 2017 of 10.7%. The system of selfassessment,
investigation, inspection and supervision by the Account Representative still shows
weaknesses with no detection of the value of the asset declaration. This study uses the theory of
planned behavior to explain the non-compliance of taxpayers. In controlling the non-compliance,
the Directorate General of Taxes can use information data that will be compared with Taxpayer
Tax Returns whose data can be obtained based on Law Number 9 of 2017 and Government
Regulation Number 31 of 2012. This study was prepared with descriptive methods by conducting
interviews, system comparisons existing with those to be submitted, and research on applicable
laws and regulations. This research shows the need for more detailed supervision of taxpayers, so
that non-compliance of taxpayers in making payments and tax reporting can be minimized.