Description
TL;DR: A small statistical forecasting exercise that fits a weighted moving average to hourly Bitcoin data and evaluates error with MAPE.
About This Document
Nashirah Abu Bakar and Sofian Rosbi use hourly observations from December 14 through December 18, 2017 to forecast the Bitcoin exchange rate. They describe the weighted moving-average method, compare predicted and observed values, and report a mean absolute percentage error for their chosen window. The paper's original title calls Bitcoin a share price, but the analysis concerns an exchange rate.
Why Bitcoin People May Care
Five days from an extraordinary bull-market period cannot establish a stable trading method. A low in-sample or short-horizon error can disappear when volatility, market structure, or the forecast window changes. This paper is useful as a compact historical methodology example, not as a live signal, investment recommendation, or evidence that weighted moving averages reliably predict Bitcoin.
What You Receive
Purchase reveals the verified preserved PDF location for this document.
Document Details
- Title: Weighted Moving Average Forecasting for Bitcoin High-Frequency Data
- Author / organization: Nashirah Abu Bakar and Sofian Rosbi
- Year: 2018
- Language: English
- Document type: Statistical forecasting research paper
- Pages: 6
CacheRat provides researched source-location and document-identification information. CacheRat does not claim ownership of the underlying third-party documentation.
