“…In predicting one step-ahead weekly gold price changes Parisi et al (2008) find that ANN has an accuracy of 61%. Using stacking and hybrid bagging Mahato and Attar (2014) find one day-ahead gold and silver price accuracy of 85% and 79%, respectively. Unlike Pierdzioch et al (2015) and Pierdzioch et al (2016a) who find that gold price trading signals generated from regression boosting offer little to no improvement over a buy and hold strategy, the results of this present paper indicate a trading strategy based on RFs offers a substantial improvement over a buy and hold strategy.…”
Section: Resultsmentioning
confidence: 99%
“…Given the interest in gold as an asset it is not surprising that there are many studies that forecast the price of gold. Examples of methods used to forecast gold prices include econometrics (Shafiee and Topal 2010;Aye et al 2015;Hassani et al 2015;Gangopadhyay et al 2016), artificial neural networks (Kristjanpoller and Minutolo 2015;Alameer et al 2019;Parisi et al 2008), boosting (Pierdzioch et al 2015(Pierdzioch et al , 2016a(Pierdzioch et al , 2016b, random forests (Liu and Li 2017;Pierdzioch and Risse 2020), support vector machines (Risse 2019), and other machine learning methods (Yazdani-Chamzini et al 2012;Livieris et al 2020;Mahato and Attar 2014).…”
Section: Introductionmentioning
confidence: 99%
“…They find that ANFIS outperforms the other models and the results are robust to different training and test sets. Mahato and Attar (2014) predict gold prices using ensemble methods. Using stacking and hybrid bagging they find gold and silver price accuracy of 85% and 79%, respectively.…”
Gold is often used by investors as a hedge against inflation or adverse economic times. Consequently, it is important for investors to have accurate forecasts of gold prices. This paper uses several machine learning tree-based classifiers (bagging, stochastic gradient boosting, random forests) to predict the price direction of gold and silver exchange traded funds. Decision tree bagging, stochastic gradient boosting, and random forests predictions of gold and silver price direction are much more accurate than those obtained from logit models. For a 20-day forecast horizon, tree bagging, stochastic gradient boosting, and random forests produce accuracy rates of between 85% and 90% while logit models produce accuracy rates of between 55% and 60%. Stochastic gradient boosting accuracy is a few percentage points less than that of random forests for forecast horizons over 10 days. For those looking to forecast the direction of gold and silver prices, tree bagging and random forests offer an attractive combination of accuracy and ease of estimation. For each of gold and silver, a portfolio based on the random forests price direction forecasts outperformed a buy and hold portfolio.
“…In predicting one step-ahead weekly gold price changes Parisi et al (2008) find that ANN has an accuracy of 61%. Using stacking and hybrid bagging Mahato and Attar (2014) find one day-ahead gold and silver price accuracy of 85% and 79%, respectively. Unlike Pierdzioch et al (2015) and Pierdzioch et al (2016a) who find that gold price trading signals generated from regression boosting offer little to no improvement over a buy and hold strategy, the results of this present paper indicate a trading strategy based on RFs offers a substantial improvement over a buy and hold strategy.…”
Section: Resultsmentioning
confidence: 99%
“…Given the interest in gold as an asset it is not surprising that there are many studies that forecast the price of gold. Examples of methods used to forecast gold prices include econometrics (Shafiee and Topal 2010;Aye et al 2015;Hassani et al 2015;Gangopadhyay et al 2016), artificial neural networks (Kristjanpoller and Minutolo 2015;Alameer et al 2019;Parisi et al 2008), boosting (Pierdzioch et al 2015(Pierdzioch et al , 2016a(Pierdzioch et al , 2016b, random forests (Liu and Li 2017;Pierdzioch and Risse 2020), support vector machines (Risse 2019), and other machine learning methods (Yazdani-Chamzini et al 2012;Livieris et al 2020;Mahato and Attar 2014).…”
Section: Introductionmentioning
confidence: 99%
“…They find that ANFIS outperforms the other models and the results are robust to different training and test sets. Mahato and Attar (2014) predict gold prices using ensemble methods. Using stacking and hybrid bagging they find gold and silver price accuracy of 85% and 79%, respectively.…”
Gold is often used by investors as a hedge against inflation or adverse economic times. Consequently, it is important for investors to have accurate forecasts of gold prices. This paper uses several machine learning tree-based classifiers (bagging, stochastic gradient boosting, random forests) to predict the price direction of gold and silver exchange traded funds. Decision tree bagging, stochastic gradient boosting, and random forests predictions of gold and silver price direction are much more accurate than those obtained from logit models. For a 20-day forecast horizon, tree bagging, stochastic gradient boosting, and random forests produce accuracy rates of between 85% and 90% while logit models produce accuracy rates of between 55% and 60%. Stochastic gradient boosting accuracy is a few percentage points less than that of random forests for forecast horizons over 10 days. For those looking to forecast the direction of gold and silver prices, tree bagging and random forests offer an attractive combination of accuracy and ease of estimation. For each of gold and silver, a portfolio based on the random forests price direction forecasts outperformed a buy and hold portfolio.
“…The work could also be extended by further validating the use of methodology in a broader set of countries and considering new sets of economic and financial variables. Moreover, the work could be validated using different forecasting horizons, spans of data sets, and complementary machine-learning strategies such as bagging and boosting techniques (Alpaydin, 2010;Mahato & Attar, 2014). Also, other econometric and data-driven models could be evaluated such as Vector Auto Regressions (VARs) and variants (such as FAVAR) (Bernanke et al, 2005) that are expected to perform very well in this type of task.…”
The opinions and results of this work are the sole responsibility of the authors. They do not represent in any way the institutional views or policies of their affiliations.
“…In [5] used analysis to determine the relationship between the gold price and various factors that influence it, such as the stock market, inflation, and interest rate. The best accuracy is found in the stacking performance model in [6]. Gold's price predictability is essential in various fields, including economics, trading, and investing.…”
This work aims to forecast gold prices for future dates using FBprophet and Linear Regression. For predicting the gold price using Linear Regression with a sample size of 140, FBprophet for time series analysis was suggested. The Dickey-Fuller test extracts seasonality (non-stationary) data and converts it to static data. The accuracy of FBProphet is 97.2 percent, compared to 85.6 percent for linear regression. Compared to linear regression, FBProphet tends to do substantially better than linear regression, with a significance level of (p<0.05). FBProphet can help predict the percentage of gold rate with greater precision.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.