We study in this paper the consequences of using the Mean Absolute Percentage Error (MAPE) as a measure of quality for regression models. We show that finding the best model under the MAPE is equivalent to doing weighted Mean Absolute Error (MAE) regression. We show that universal consistency of Empirical Risk Minimiza…
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
169,291 papers · 148 categories
Trend · papers per month
48 results for “Mean Absolute Percentage Error”
We study in this paper the consequences of using the Mean Absolute Percentage Error (MAPE) as a measure of quality for regression models. We show that finding the best model under the MAPE is equivalent to doing weighted Mean Absolute Error (MAE) regression. We also show that, under some asumptions, universal consisten…
We study in this paper the consequences of using the Mean Absolute Percentage Error (MAPE) as a measure of quality for regression models. We prove the existence of an optimal MAPE model and we show the universal consistency of Empirical Risk Minimization based on the MAPE. We also show that finding the best model under…
Forecasting Exchange Rates Using Time Series Analysis: The sample of the currency of Kazakhstanq-fin.ST
This paper models yearly exchange rates between USD/KZT, EUR/KZT and SGD/KZT, and compares the actual data with developed forecasts using time series analysis over the period from 2006 to 2014. The official yearly data of National Bank of the Republic of Kazakhstan is used for present study. The main goal of this paper…
Neural networks predict stock prices better than traditional methods.
problem Predicting stock prices in volatile financial markets.
method Compared five neural network models (BP, RBF, GRNN, SVMR, LS-SVMR) on three stocks.
result BP neural network outperformed other models in accuracy.
An Exponential Factorization Machine with Percentage Error Minimization to Retail Sales Forecastingcs.LG
New model predicts sales of new products with short life cycles.
problem Forecasting sales of new products with short lead times and life cycles.
method Developed an exponential factorization machine (EFM) to consider attributes and pairwise interactions.
result EFM model outperforms existing models in terms of MAPE and MAE.
Research evaluates ANN stock price prediction system for Shanghai Stock Exchange.
problem Predicting exact stock prices in the Shanghai Stock Exchange.
method Feedforward multi-layer perceptron with error backpropagation, using 5:21:21:1 configuration with 80% training data.
result Neural networks can predict stock prices with low mean absolute percentage errors (1.95%).
Research uses Sutte Indicator to predict stock market movements.
problem Predicting stock market movements accurately.
method Applied Sutte Indicator alongside SMA and MACD for comparison.
result Sutte Indicator shows better reliability in predicting stock movements.
Predicts Bitcoin price using Twitter sentiment analysis.
problem Volatility and varied opinions in cryptocurrency markets.
method Developed a model combining sentiment analysis of tweets and historical price data.
result Sentiment prediction MAPE of 9.45%, price prediction MAPE of 3.6%
New features from early battery cycles predict lifetime with high accuracy.
problem Accurately predicting battery lifetime under varying conditions is challenging due to manufacturing variability and usage-dependent degradation.
method Extracted features from regularly scheduled reference performance tests and used them to predict battery lifetime using a hierarchical Bayesian regression model.
result Demonstrated a lifetime prediction of in-distribution cells with 15.1% mean absolute percentage error using only the first 15% of data.
Hinge-FM2I fills missing data in time series with high accuracy.
problem Handling missing data in univariate time series.
method Inspired by door hinges, Hinge-FM2I imputes missing data using FM2I and selects the best imputed gap.
result Hinge-FM2I significantly outperforms other methods in sMAPE scores.
We have applied a Long Short-Term Memory neural network to model S&P 500 volatility, incorporating Google domestic trends as indicators of the public mood and macroeconomic factors. In a held-out test set, our Long Short-Term Memory model gives a mean absolute percentage error of 24.2%, outperforming linear Ridge/Lasso…
Predict stock prices using financial news sentiment analysis.
problem Predicting stock market trends for better investment returns.
method Deep Learning (MLP, LSTM, FinBERT-LSTM) integrating news sentiment.
result FinBERT-LSTM model predicts stock prices more accurately.
Are Bitcoins price predictable? Evidence from machine learning techniques using technical indicatorsq-fin.ST
Machine learning predicts Bitcoin price with high accuracy.
problem Uncertainty in Bitcoin price prediction for investors.
method Used machine learning techniques with technical indicators.
result Stacking ensemble model with random forest and GLM is optimal.
The study forecasts portfolio volatility using cointegrated asset dynamics.
problem Forecasting volatility in portfolios with high accuracy.
method Developed HVR/DVR ratios and used Vector Error Correction Model (VECM) to forecast volatility.
result VECM forecasts of portfolio volatility have lower MAPE than covariance-based forecasts.
The paper bounds the mean absolute error in DNN vector-to-vector regression.
problem Bounding the mean absolute error in deep neural network based vector-to-vector regression.
method Error decomposition techniques in statistical learning theory and non-convex optimization theory were used to derive upper bounds for approximation, estimation, and optimization errors.
result Theoretical upper bounds for mean absolute error in DNN vector-to-vector regression were derived and validated experimentally.
The study improves heat demand prediction using neural networks and weather parameters.
problem Improving accuracy of heat demand prediction in district heating networks.
method Used an Elman neural network to investigate the impact of direct solar irradiance and wind speed on heat demand prediction.
result Including wind speed generally results in lower MAPE than including direct solar irradiance, but direct solar irradiance achieves lower maximum absolute deviation.
Study evaluates 41 ML models for Bitcoin trading performance.
problem Predicting Bitcoin prices for algorithmic trading.
method Examined 21 classifiers and 20 regressors under various market conditions.
result Certain models like Random Forest and Stochastic Gradient Descent outperform others in profit and risk management.
Hybrid model outperforms benchmarks in financial forecasting.
problem Robust asset price forecasting in finance.
method Combining LSTM with Neural Levy Processes using Grey Wolf Optimizer and ANN calibration.
result Hybrid model outperforms base LSTM and other models.
Deep model improves option pricing for CSI 300 index with sentiment and volatility features.
problem Challenges in real market option pricing, especially with constant volatility assumption.
method Deep Forward-Backward Stochastic Differential Equation (FBSDE) framework with dual-network architecture.
result Significant reduction in MAE and MAPE compared to BSM model.
Model predicts passenger origin-destination for online taxi-hailing systems.
problem Predicting passenger origin-destination for efficient transportation planning.
method K-means clustering, non-negative matrix factorization, stacked recurrent neural network.
result Proposed model reduces MAPE by 5-7% for 1-hour windows and 14% for 30-minute windows.
Hybrid model predicts stock prices more accurately.
problem Efficient stock price prediction.
method Symbiotic organisms search algorithm trained feedforward neural networks.
result Outstanding predictive performance compared to other models.
Comparative Analysis of Machine Learning Approaches to Analyze and Predict the Covid-19 Outbreakstat.ML
This paper compares ML models for predicting COVID-19 trends.
problem Forecasting the spread of COVID-19 to reduce its impact.
method Applied ARDL method to identify relationships, then used ML models (SVM, RF, KNN, ANN) for prediction.
result Models accurately forecasted COVID-19 cases with low MAPE.
Unified ML approach predicts ED attendances with high accuracy.
problem Managing hospital demand at emergency departments efficiently.
method Ensemble of time series and machine learning approaches with hyperparameter tuning.
result Predictions with mean absolute error of +/- 14 and +/- 10 patients, MAE of 6.8% and 8.6%.
Predicts fine-grained OD matrices for ridesharing platforms to optimize supply-demand balance.
problem Accurately predicting spatial-temporal OD demands for ridesharing platforms.
method OD-CED model combining unsupervised space coarsening and encoder-decoder architecture.
result Significant improvement in prediction accuracy (45% RMSE reduction, 60% WAPE reduction).
Paper improves language models' ability to predict numbers.
problem Improving language models' numeracy for technical documents.
method Exploring memorisation, digit-by-digit composition, and a continuous probability density function model.
result Hierarchical models improve perplexity by 2 and 4 orders of magnitude.
Meta-learning predicts optimal ensemble size and methods for time series forecasting.
problem Finding the best ensemble of time series forecasting methods.
method Two-step approach using meta-learning to predict ensemble size and methods.
result Meta-learning outperformed benchmarks in forecasting errors for all data types and horizons.
The paper explores MAE as a loss function for DNN vector-to-vector regression, proving its advantages over MSE.
problem Improving loss function for deep neural network based vector-to-vector regression.
method Presenting performance bounds and new properties of MAE, deriving generalized upper bounds, and interpreting MAE as a Laplacian distribution.
result MAE is a more suitable loss function than MSE for DNN based vector-to-vector regression, especially when errors follow a Laplacian distribution.
This study compares deep learning and statistical models for stock price forecasting.
problem Accurate stock price prediction is challenging due to market volatility.
method Used deep learning (LSTM, RNN, CNN, FULL CNN) and statistical models (ARIMA, Moving Averages) on S&P 500 data.
result LSTM model showed the lowest Mean Absolute Error (MAE), indicating highest accuracy.
SAGA predicts multi-year earnings with adaptive intervals, improving forecast accuracy.
problem Forecasting long-range nonlinear structure in lifetime earnings.
method Decoder-only transformer for irregular tabular sequences, split conformal calibration.
result Significant improvement in forecast accuracy compared to existing methods.
New online adaptive SVR model for IVS with hardware acceleration.
problem Modeling implied volatility surface (IVS) in real-time.
method Online adaptive primal support vector regression (SVR) with hardware acceleration.
result Gaussian kernel outperforms linear kernel in support vector size regulation.
Option Pricing of Twin Assetsq-fin.PR
How to price and hedge claims on nontraded assets are becoming increasingly important matters in option pricing theory today. The most common practice to deal with these issues is to use another similar or "closely related" asset or index which is traded, for hedging purposes. Implicitly, traders assume here that the h…
Study forecasts vegetable prices in Nepal using a novel index and ensemble model.
problem High volatility and cultural influences on agricultural commodity prices.
method Developed KVPI, created features, evaluated multiple models, introduced Momentum-Corrected Online Stacking Ensemble.
result Achieved RMSE of 1.771, MAPE of 0.68%, and R-squared of 0.845 at 90-day horizon.
Optimal stock price prediction model using recurrent neural networks with RMSprop optimizer.
problem Stock price prediction using neural networks.
method Comparison of fully connected, convolutional, and recurrent architectures; inclusion of three optimization techniques.
result Single layer recurrent neural network with RMSprop optimizer produces optimal results with validation and test MAE of 0.0150 and 0.0148 respectively.
An Estimation of Personnel Food Demand Quantity for Businesses by Using Artificial Neural Networkscs.LG
Study uses AI to estimate food demand for businesses.
problem Difficulty in predicting daily food demand for personnel.
method Artificial Neural Networks (ANN) model trained on historical data.
result ANN model predicts daily food demand with low error rate.
Method estimates parameters of complex nonlinear systems.
problem Parameter estimation for nonlinear systems with derivative states.
method Regularized linear regression using differentiation filtering and least squares.
result Finite-sample bound on mean absolute error of estimation.
Two heuristics solve dynamic multiple travelling salesmen problems.
problem Dynamic routing with unknown customers.
method Balanced dynamic closest vehicle heuristic and balanced dynamic assignment vehicle heuristic.
result Continuous approximation models for strategic dynamic routing.
LALR adapts learning rate for faster convergence in regression and neural nets.
problem Finding optimal learning rates for faster convergence in regression and neural networks.
method Lipschitz continuity theory applied to Mean Absolute Error and Quantile loss functions.
result Adaptive learning rate policy enables up to 20x faster convergence.
Enhances renewable energy forecasts and trading profits.
problem Improving accuracy and economic benefits of renewable energy forecasts and trading.
method Data-driven newsvendor model using features from multiple sources.
result Significant reduction in balancing costs and improvement in forecast accuracy.
Extrapolation to complete basis-set limit in density-functional theory by quantile random-forest modelsphysics.comp-ph
Machine learning model predicts DFT total energy to complete basis set limit.
problem Finding a model to extrapolate DFT calculations to complete basis set limit.
method Quantile-random-forest model trained on binary solids data.
result Random-forest model achieves <25% symmetric MAPE for both DFT codes.
Paper proposes a new metric to evaluate survival models, especially for censored data.
problem Challenges in evaluating survival prediction models due to censored data.
method Developed a novel approach to estimate Mean Absolute Error (MAE) for survival datasets with censored data.
result The proposed MAE metric using pseudo-observations accurately ranks model performance and closely matches true MAE.
Machine learning predicts band gaps for large organic crystals.
problem Predicting band gaps for complex organic crystal structures.
method Released a dataset of 12,500 crystal structures and their band gaps. Trained two state-of-the-art models to achieve a mean absolute error of 0.388 eV.
result Trained models predict band gaps with 13% error for an average gap of 3.05 eV.
Study shows accuracy of neural networks depends more on error location than percentage of error.
problem Effect of noise on accuracy in incremental learning neural networks.
method Empirical study using Perceptron, Feed Forward Neural Network, and Radial Basis Function Neural Network.
result Accuracy of neural networks is more dependent on error location than the percentage of error.
It is well known that the distribution of returns from various financial instruments are leptokurtic, meaning that the distributions have "fatter tails" than a Normal distribution, and have skew toward zero. This paper presents a graceful micro-level explanation for such fat-tailed outcomes, using agents whose private …
MetaCI uses meta-learning to generalize causal inference across heterogeneous populations.
problem Generalizing causal inference from multiple homogeneous subgroups to a heterogeneous population.
method MetaCI employs meta-learning to handle data distribution shifts and concept drifts.
result MetaCI outperforms random initialization and other methods in counterfactual inference.
A hybrid physics-ML model predicts FO water flux with high accuracy and uncertainty quantification.
problem Challenges in accurately modeling Forward Osmosis water flux due to complex internal mass transfer phenomena.
method Robust Hybrid Physics-ML framework using Gaussian Process Regression (GPR) for uncertainty-aware Jw prediction.
result Achieved a state-of-the-art MAPE of 0.26% and R2 of 0.999 on independent test data.
This study improves sales forecasting for Intel Corporation in the semiconductor industry.
problem Accurate sales forecasting in the semiconductor industry for Intel Corporation.
method Innovative incorporation of various indicators into quantitative models, including multiple regressions, time series analysis, random forest, and boosting tree. Ensemble models selected based on validation errors and moving windows validation.
result Development of an ensemble model that captures distinct characteristics at lead time and lines of business levels, improving response to market fluctuations.
This paper achieves optimal regret bounds for locally private linear contextual bandit.
problem Designing locally private linear contextual bandit algorithms with optimal regret bounds.
method New algorithmic and analytical ideas, including mean absolute deviation analysis and layered principal component regression.
result Achieves an regret upper bound for locally private linear contextual bandit.