Scribd Logo
Upload

Welcome to Scribd!

  • 23 views

    Economics of Gold Price Movement-Forecasting Analysis Using Macro-Economic, Investor Fear and Investor Behavior Features

    Uploaded by

    Speculation Only
    This document discusses using macroeconomic factors, investor fear features, and investor behavior features to analyze and forecast gold prices. It analyzes gold prices at both monthly and weekly resolutions using factors like CPI, forex rates, stress indexes, Twitter sentiment, and search volume indexes. Statistical techniques like correlation, cross-correlation, and Granger causality are used to establish relationships between gold prices and other factors. Forecasting techniques like EMMS and SVM are also used to predict gold prices and movement direction. The goal is to provide a more comprehensive analysis of how both fundamental economic factors and behavioral investor perceptions impact gold prices.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Economics of Gold Price Movement-Forecasting Analysis Using Macro-Economic, Investor Fear and Investor Behavior Features

    Uploaded by

    Speculation Only
    23 views2 pages
    This document discusses using macroeconomic factors, investor fear features, and investor behavior features to analyze and forecast gold prices. It analyzes gold prices at both monthly and weekly resolutions using factors like CPI, forex rates, stress indexes, Twitter sentiment, and search volume indexes. Statistical techniques like correlation, cross-correlation, and Granger causality are used to establish relationships between gold prices and other factors. Forecasting techniques like EMMS and SVM are also used to predict gold prices and movement direction. The goal is to provide a more comprehensive analysis of how both fundamental economic factors and behavioral investor perceptions impact gold prices.

    Original Description:

    DATA FX PART1

    Original Title

    978-3-642-35542-4_10

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses using macroeconomic factors, investor fear features, and investor behavior features to analyze and forecast gold prices. It analyzes gold prices at both monthly and weekly resolutions using factors like CPI, forex rates, stress indexes, Twitter sentiment, and search volume indexes. Statistical techniques like correlation, cross-correlation, and Granger causality are used to establish relationships between gold prices and other factors. Forecasting techniques like EMMS and SVM are also used to predict gold prices and movement direction. The goal is to provide a more comprehensive analysis of how both fundamental economic factors and behavioral investor perceptions impact gold prices.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    23 views2 pages

    Economics of Gold Price Movement-Forecasting Analysis Using Macro-Economic, Investor Fear and Investor Behavior Features

    Uploaded by

    Speculation Only
    This document discusses using macroeconomic factors, investor fear features, and investor behavior features to analyze and forecast gold prices. It analyzes gold prices at both monthly and weekly resolutions using factors like CPI, forex rates, stress indexes, Twitter sentiment, and search volume indexes. Statistical techniques like correlation, cross-correlation, and Granger causality are used to establish relationships between gold prices and other factors. Forecasting techniques like EMMS and SVM are also used to predict gold prices and movement direction. The goal is to provide a more comprehensive analysis of how both fundamental economic factors and behavioral investor perceptions impact gold prices.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    Economics of Gold Price Movement-Forecasting

    Analysis Using Macro-economic, Investor Fear


    and Investor Behavior Features

    Jatin Kumar1 , Tushar Rao2 , and Saket Srivastava1


    1
    IIIT-Delhi
    2
    Netaji Subhas Institute of Technology- Delhi
    {saket,jatin09021}@iiitd.ac.in, [email protected]

    Abstract. Recent works have shown that besides fundamental factors


    like interest rate, inflation index and foreign exchange rates, behavioral
    factors like consumer sentiments and global economic stability play an
    important role in driving gold prices at shorter time resolutions. In this
    work we have done comprehensive modeling of price movements of gold,
    using three feature sets, namely- macroeconomic factors (using CPI in-
    dex and foreign exchange rates), investor fear features (using US Econ-
    omy Stress Index and gold ETF Volatility Index) and investor behavior
    features (using the sentiment of Twitter feeds and web search volume
    index from Google Search Volume Index). Our results bring insights like
    high correlation (upto 0.92 for CPI) between various features, which is
    a significant improvement over earlier works. Using Grangers causality
    analysis, we have validated that the movement in gold price is greatly
    affected in the short term by some features, consistently over a five week
    lag. Finally, we implemented forecasting techniques like expert model
    mining system (EMMS) and binary SVM classifier to demonstrate fore-
    casting performance using different features.

    Keywords: Gold, Twitter, SVI, Stress Index, VIX, Forex rates.

    1 Introduction

    Most of the academic or industrial work on gold prices make use of only fun-
    damental macro-economic factors without taking into account the investor per-
    ceptions and sentiments towards gold and economy in general. We analyze the
    gold prices using three types of factors- macro-economic factors, factors depict-
    ing level of economic stability or investor fear features and the investor behavior
    features that we get from Twitter and Search Volume Index (Google SVI). We
    apply techniques like pearson correlation, cross-correlation at different time lags
    and Grangers causality analysis for establishing a causative relationship between
    the gold prices and the above mentioned features. Further, we use forecasting
    techniques like EMMS and SVM to predict gold prices and the direction of its
    movement. To the best of our knowledge, there is no earlier work that deals

    S. Srinivasa and V. Bhatnagar (Eds.): BDA 2012, LNCS 7678, pp. 111–121, 2012.
    
    c Springer-Verlag Berlin Heidelberg 2012
    112 J. Kumar, T. Rao, and S. Srivastava

    with the comprehensive comparison between the investor perception and senti-
    ment with macro-economic factors [5, 8–10]. This paper marks a novel step in
    the direction of analyzing the effect of gold prices on macro-economic factors in
    combination with investor’s perception or behavior. Using the macro-economic
    factors, we try to establish a long-term pattern of these factors with the gold
    prices, while we use Twitter feed data, Google SVI or gold ETF volatility index
    to study the short-term fluctuations in gold prices.
    Figure 1 summarizes the work flow of the paper. In section 2 we present data
    collection and prior processing, defining the terminologies used in the market
    securities. Further, in section 3 we present the statistical techniques implemented
    and subsequently discuss the derived results.

    Fig. 1. Work flow of the techniques implemented in this research work

    2 Dataset Preparation and Mining


    We have made use of spot gold price data from World gold Council at two levels
    of analysis- monthly for macroeconomic and investor fear features and weekly
    for investor behavior features 1 . Further, in this section we discuss the collection
    and processing of the time series data used in this work.

    2.1 Macro-economic Factors


    We use monthly time series for the macroeconomic factors - Consumer Price
    Index (CPI) and forex rates of USD with major currencies. Time period of
    analysis in this features set is from January 1st 2003 to January 1st 2012. This
    time period corresponds to one of the robust changes in the economy, including
    meteoric rise in 2004 and recession in 2008 due to sub-prime mortgage crisis.
    This provides us more confidence in the causative relations and the conclusions
    drawn from the statistical results. The CPI in the United States is defined by
    the Bureau of Labor Statistics as ”a measure of the average change over time
    1
    http://www.gold.org/investment/statistics/goldpricechart/

    You might also like