hscrown's picture
Update README.md
c9fbefe
metadata
task_categories:
  - time-series-forecasting
language:
  - en
  - ko
tags:
  - mining
  - LSTM
  - TimeSeries
  - CanariaView

CanariaView Global Copper Demand Forecasting Dataset

Description

This dataset encompasses economic and industrial indicators vital for constructing a copper demand forecasting model.

Coverage Period: Monthly data from January 1995 to March 2023, encompassing a total of 339 months.

Column Descriptions and Sources:

  • HSI_value (US Housing Starts Index): Y-Chart
  • CCI_value (Consumer Confidence Index): OECD
  • IPI_value (Industrial Production Total Index): FRED
  • GDPC_value (Real Gross Domestic Product): FRED
  • Copper price: MacroTrends

Preprocessing Methodology and Data Collection Details:

  • Comprehensive analysis of data structure followed by essential preprocessing.
  • Appropriate handling of missing values.
  • Daily and quarterly data uniformly expanded to a monthly timescale for consistency.
  • Daily data (e.g., Copper price) and quarterly data (e.g., GDPC_value)
  • Dependent variable data used in the model was available from 1995, guiding the collection of independent variables-this dataset- from that year.

한국어 설명

본 데이터셋은 구리 수요 예측 모델 구축을 위한 경제지표 및 산업지표로 구성되었습니다.

기간: 1995년 1월~2023년 3월(월별), 총 339개월.

컬럼 설명 및 출처

  • HSI_value (미국 주택착공지수): Y-Chart
  • CCI_value (미국 소비자신뢰지수): OECD
  • IPI_value (미국 산업생산자지수): FRED
  • GDPC_value (미국 실질 GDP): FRED
  • Copper price (구리 가격): MacroTrends

데이터 전처리 및 수집 방법:

  • 데이터 구조 분석 및 전처리 과정 수행.
  • 결측치 처리.
  • 일별 및 분기별 자료는 월별 데이터로의 확장을 통해 일관된 시계열 데이터로 통합.
  • 일별 자료 (구리 가격), 분기별 자료 (GDPC_value)
  • 수요 모델에 사용된 종속변수 데이터가 1995년부터 확보되어 독립변수인 본 데이터셋도 1995년도부터 수집함.