In this article, we present a method to forecast the Portuguese gross domestic product (GDP) in each current quarter (nowcasting). It combines bridge equations of the real GDP on readily available monthly data like the Economic Sentiment Indicator (ESI), industrial production index, cement sales or exports and imports, with forecasts for the jagged missing values computed with the well-known Hodrick and Prescott (HP) filter. As shown, this simple multivariate approach can perform as well as a Targeted Diffusion Index (TDI) model and slightly better than the univariate Theta method in terms of out-of-sample mean errors.
Healthcare datasets obtained from Electronic Health Records have proven to be extremely useful to assess associations between patients' predictors and outcomes of interest. However, these datasets often suffer from missing values in a high proportion of cases and the simple removal of these cases may introduce severe bias. For these reasons, several multiple imputation algorithms have been proposed to attempt to recover the missing information. Each algorithm presents strengths and weaknesses, and there is currently no consensus on which multiple imputation algorithms works best in a given scenario. Furthermore, the selection of each algorithm parameters and data-related modelling choices are also both crucial and challenging. In this paper, we propose a novel framework to numerically evaluate strategies for handling missing data in the context of statistical analysis, with a particular focus on multiple imputation techniques. We demonstrate the feasibility of our approach on a large cohort of type-2 diabetes patients provided by the National COVID Cohort Collaborative (N3C) Enclave, where we explored the influence of various patient characteristics on outcomes related to COVID-19. Our analysis included classic multiple imputation techniques as well as simple complete-case Inverse Probability Weighted models. The experiments presented here show that our approach could effectively highlight the most valid and performant missing-data handling strategy for our case study. Moreover, our methodology allowed us to gain an understanding of the behavior of the different models and of how it changed as we modified their parameters. Our method is general and can be applied to different research fields and on datasets containing heterogeneous types.
While data are the primary fuel for machine learning models, they often suffer from missing values, especially when collected in real-world scenarios. However, many off-the-shelf machine learning models, including artificial neural network models, are unable to handle these missing values directly. Therefore, extra data preprocessing and curation steps, such as data imputation, are inevitable before learning and prediction processes. In this study, we propose a simple and intuitive yet effective method for pruning missing values (PROMISSING) during learning and inference steps in neural networks. In this method, there is no need to remove or impute the missing values; instead, the missing values are treated as a new source of information (representing what we do not know). Our experiments on simulated data, several classification and regression benchmarks, and a multi-modal clinical dataset show that PROMISSING results in similar prediction performance compared to various imputation techniques. In addition, our experiments show models trained using PROMISSING techniques are becoming less decisive in their predictions when facing incomplete samples with many unknowns. This finding hopefully advances machine learning models from being pure predicting machines to more realistic thinkers that can also say "I do not know" when facing incomplete sources of information.
The widespread availability of satellite images has allowed researchers to model complex systems such as disease dynamics. However, many satellite images have missing values due to measurement defects, which render them unusable without data imputation. For example, the scanline corrector for the LANDSAT 7 satellite broke down in 2003, resulting in a loss of around 20\% of its data. Inpainting involves predicting what is missing based on the known pixels and is an old problem in image processing, classically based on PDEs or interpolation methods, but recent deep learning approaches have shown promise. However, many of these methods do not explicitly take into account the inherent spatiotemporal structure of satellite images. In this work, we cast satellite image inpainting as a natural meta-learning problem, and propose using convolutional neural processes (ConvNPs) where we frame each satellite image as its own task or 2D regression problem. We show ConvNPs can outperform classical methods and state-of-the-art deep learning inpainting models on a scanline inpainting problem for LANDSAT 7 satellite images, assessed on a variety of in and out-of-distribution images.
In this paper, we introduce a probabilistic model for learning nonnegative matrix factorization (NMF) that is commonly used for predicting missing values and finding hidden patterns in the data, in which the matrix factors are latent variables associated with each data dimension. The nonnegativity constraint for the latent factors is handled by choosing priors with support on the nonnegative subspace. Bayesian inference procedure based on Gibbs sampling is employed. We evaluate the model on several real-world datasets including MovieLens 100K and MovieLens 1M with different sizes and dimensions and show that the proposed Bayesian NMF GRRN model leads to better predictions and avoids overfitting compared to existing Bayesian NMF approaches.
Incomplete observability of data generates an identification problem. There is no panacea for missing data. What one can learn about a population parameter depends on the assumptions one finds credible to maintain. The credibility of assumptions varies with the empirical setting. No specific assumptions can provide a realistic general solution to the problem of inference with missing data. Yet Rubin has promoted random multiple imputation (RMI) as a general way to deal with missing values in public-use data. This recommendation has been influential to empirical researchers who seek a simple fix to the nuisance of missing data. This paper adds to my earlier critiques of imputation. It provides a transparent assessment of the mix of Bayesian and frequentist thinking used by Rubin to argue for RMI. It evaluates random imputation to replace missing outcome or covariate data when the objective is to learn a conditional expectation. It considers steps that might help combat the allure of making stuff up.
Data values in a dataset can be missing or anomalous due to mishandling or human error. Analysing data with missing values can create bias and affect the inferences. Several analysis methods, such as principle components analysis or singular value decomposition, require complete data. Many approaches impute numeric data and some do not consider dependency of attributes on other attributes, while some require human intervention and domain knowledge. We present a new algorithm for data imputation based on different data type values and their association constraints in data, which are not handled currently by any system. We show experimental results using different metrics comparing our algorithm with state of the art imputation techniques. Our algorithm not only imputes the missing values but also generates human readable explanations describing the significance of attributes used for every imputation.
Genomics data such as RNA gene expression, methylation and micro RNA expression are valuable sources of information for various clinical predictive tasks. For example, predicting survival outcomes, cancer histology type and other patients' related information is possible using not only clinical data but molecular data as well. Moreover, using these data sources together, for example in multitask learning, can boost the performance. However, in practice, there are many missing data points which leads to significantly lower patient numbers when analysing full cases, which in our setting refers to all modalities being present. In this paper we investigate how imputing data with missing values using deep learning coupled with multitask learning can help to reach state-of-the-art performance results using combined genomics modalities, RNA, micro RNA and methylation. We propose a generalised deep imputation method to impute values where a patient has all modalities present except one. Interestingly enough, deep imputation alone outperforms multitask learning alone for the classification and regression tasks across most combinations of modalities. In contrast, when using all modalities for survival prediction we observe that multitask learning alone outperforms deep imputation alone with statistical significance (adjusted p-value 0.03). Thus, both approaches are complementary when optimising performance for downstream predictive tasks.
Albeit of crucial interest for both financial practitioners and researchers, market-implied volatility data of European swaptions often exhibit large portions of missing quotes due to illiquidity of the various underlying swaption instruments. In this case, standard stochastic interpolation tools like the common SABR model often cannot be calibrated to observed implied volatility smiles, due to data being only available for the at-the-money quote of the respective underlying swaption. Here, we propose to infer the geometry of the full unknown implied volatility cube by learning stochastic latent representations of implied volatility cubes via variational autoencoders, enabling inference about the missing volatility data conditional on the observed data by an approximate Gibbs sampling approach. Imputed estimates of missing quotes can afterwards be used to fit a standard stochastic volatility model. Since training data for the employed variational autoencoder model is usually sparsely available, we test the robustness of the approach for a model trained on synthetic data on real market quotes and we show that SABR interpolated volatilites calibrated to reconstructed volatility cubes with artificially imputed missing values differ by not much more than two basis points compared to SABR fits calibrated to the complete cube. Moreover, we show how the imputation can be used to successfully set up delta-neutral portfolios for hedging purposes.
Urban traffic flow prediction using data-driven models can play an important role in route planning and preventing congestion on highways. These methods utilize data collected from traffic recording stations at different timestamps to predict the future status of traffic. Hence, data collection, transmission, storage, and extraction techniques can have a significant impact on the performance of the traffic flow model. On the other hand, a comprehensive database can provide the opportunity for using complex, yet reliable predictive models such as deep learning methods. However, most of these methods have difficulties in handling missing values and outliers. This study focuses on hybrid deep neural networks to predict traffic flow in the California Freeway Performance Measurement System (PeMS) with missing values. The proposed networks are based on a combination of recurrent neural networks (RNNs) to consider the temporal dependencies in the data recorded in each station and convolutional neural networks (CNNs) to take the spatial correlations in the adjacent stations into account. Various architecture configurations with series and parallel connections are considered based on RNNs and CNNs, and several prevalent data imputation techniques are used to examine the robustness of the hybrid networks to missing values. A comprehensive analysis performed on two different datasets from PeMS indicates that the proposed series-parallel hybrid network with the mean imputation technique achieves the lowest error in predicting the traffic flow and is robust to missing values up until 21% missing ratio in both complete and incomplete training data scenarios when applied to an incomplete test data.
Graph-based spatio-temporal neural networks are effective to model the spatial dependency among discrete points sampled irregularly from unstructured grids, thanks to the great expressiveness of graph neural networks. However, these models are usually spatially-transductive -- only fitting the signals for discrete spatial nodes fed in models but unable to generalize to `unseen' spatial points with zero-shot. In comparison, for forecasting tasks on continuous space such as temperature prediction on the earth's surface, the \textit{spatially-inductive} property allows the model to generalize to any point in the spatial domain, demonstrating models' ability to learn the underlying mechanisms or physics laws of the systems, rather than simply fit the signals. Besides, in temporal domains, \textit{irregularly-sampled} time series, e.g. data with missing values, urge models to be temporally-continuous. Motivated by the two issues, we propose a spatio-temporal framework based on neural operators for PDEs, which learn the underlying mechanisms governing the dynamics of spatially-continuous physical quantities. Experiments show our model's improved performance on forecasting spatially-continuous physic quantities, and its superior generalization to unseen spatial points and ability to handle temporally-irregular data.
Missing values are common in many real-life datasets. However, most of the current machine learning methods can not handle missing values. This means that they should be imputed beforehand. Gaussian Processes (GPs) are non-parametric models with accurate uncertainty estimates that combined with sparse approximations and stochastic variational inference scale to large data sets. Sparse GPs can be used to compute a predictive distribution for missing data. Here, we present a hierarchical composition of sparse GPs that is used to predict missing values at each dimension using all the variables from the other dimensions. We call the approach missing GP (MGP). MGP can be trained simultaneously to impute all observed missing values. Specifically, it outputs a predictive distribution for each missing value that is then used in the imputation of other missing values. We evaluate MGP in one private clinical data set and four UCI datasets with a different percentage of missing values. We compare the performance of MGP with other state-of-the-art methods for imputing missing values, including variants based on sparse GPs and deep GPs. The results obtained show a significantly better performance of MGP.
In this study we applied the CART-type Decision Tree (DT-CART) method derived from artificial intelligence technique to the prediction of the solvency of bank customers, for this we used historical data of bank customers. However we have adopted the process of Data Mining techniques, for this purpose we started with a data preprocessing in which we clean the data and we deleted all rows with outliers or missing values as well as rows with empty columns, then we fixed the variable to be explained (dependent or Target) and we also thought to eliminate all explanatory (independent) variables that are not significant using univariate analysis as well as the correlation matrix, then we applied our CART decision tree method using the SPSS tool. After completing our process of building our model (AD-CART), we started the process of evaluating and testing the performance of our model, by which we found that the accuracy and precision of our model is 71%, so we calculated the error ratios, and we found that the error rate equal to 29%, this allowed us to conclude that our model at a fairly good level in terms of precision, predictability and very precisely in predicting the solvency of our banking customers.
Tensor factorization models are widely used in many applied fields such as chemometrics, psychometrics, computer vision or communication networks. Real life data collection is often subject to errors, resulting in missing data. Here we focus in understanding how this issue should be dealt with for nonnegative tensor factorization. We investigate several criteria used for non-negative tensor factorization in the case where some entries are missing. In particular we show how smoothness penalties can compensate the presence of missing values in order to ensure the existence of an optimum. This lead us to propose new criteria with efficient numerical optimization algorithms. Numerical experiments are conducted to support our claims.
Movement speed data from urban road networks, computed from ridesharing vehicles or taxi trajectories, is often high-dimensional, sparse, and nonstationary (e.g., exhibiting seasonality). To address these challenges, we propose a Nonstationary Temporal Matrix Factorization (NoTMF) model that leverages matrix factorization to project high-dimensional and sparse movement speed data into low-dimensional latent spaces. This results in a concise formula with the multiplication between spatial and temporal factor matrices. To characterize the temporal correlations, NoTMF takes a latent equation on the seasonal differenced temporal factors using higher-order vector autoregression (VAR). This approach not only preserves the low-rank structure of sparse movement speed data but also maintains consistent temporal dynamics, including seasonality information. The learning process for NoTMF involves optimizing the spatial and temporal factor matrices along with a collection of VAR coefficient matrices. To solve this efficiently, we introduce an alternating minimization framework, which tackles a challenging procedure of estimating the temporal factor matrix using conjugate gradient method, as the subproblem involves both partially observed matrix factorization and seasonal differenced VAR. To evaluate the forecasting performance of NoTMF, we conduct extensive experiments on Uber movement speed datasets, which are estimated from ridesharing vehicle trajectories. These datasets contain a large proportion of missing values due to insufficient ridesharing vehicles on the urban road network. Despite the presence of missing data, NoTMF demonstrates superior forecasting accuracy and effectiveness compared to baseline models. Moreover, as the seasonality of movement speed data is of great concern, the experiment results highlight the significance of addressing the nonstationarity of movement speed data.
World Input-Output (I/O) matrices provide the networks of within- and cross-country economic relations. In the context of I/O analysis, the methodology adopted by national statistical offices in data collection raises the issue of obtaining reliable data in a timely fashion and it makes the reconstruction of (part of) the I/O matrices of particular interest. In this work, we propose a method combining hierarchical clustering and Matrix Completion (MC) with a LASSO-like nuclear norm penalty, to impute missing entries of a partially unknown I/O matrix. Through simulations based on synthetic matrices we study the effectiveness of the proposed method to predict missing values from both previous years data and current data related to countries similar to the one for which current data are obscured. To show the usefulness of our method, an application based on World Input-Output Database (WIOD) tables - which are an example of industry-by-industry I/O tables - is provided. Strong similarities in structure between WIOD and other I/O tables are also found, which make the proposed approach easily generalizable to them.
Industrial machine learning systems face data challenges that are often under-explored in the academic literature. Common data challenges are data distribution shifts, missing values and anomalies. In this paper, we discuss data challenges and solutions in the context of a Neural Forecasting application on labor planning.We discuss how to make this forecasting system resilient to these data challenges. We address changes in data distribution with a periodic retraining scheme and discuss the critical importance of model stability in this setting. Furthermore, we show how our deep learning model deals with missing values natively without requiring imputation. Finally, we describe how we detect anomalies in the input data and mitigate their effect before they impact the forecasts. This results in a fully autonomous forecasting system that compares favorably to a hybrid system consisting of the algorithm and human overrides.
Systems are commonly monitored for health and security through collection and streaming of multivariate time series. Advances in time series forecasting due to adoption of multilayer recurrent neural network architectures make it possible to forecast in high-dimensional time series, and identify and classify novelties early, based on subtle changes in the trends. However, mainstream approaches to multi-variate time series predictions do not handle well cases when the ongoing forecast must include uncertainty, nor they are robust to missing data. We introduce a new architecture for time series monitoring based on combination of state-of-the-art methods of forecasting in high-dimensional time series with full probabilistic handling of uncertainty. We demonstrate advantage of the architecture for time series forecasting and novelty detection, in particular with partially missing data, and empirically evaluate and compare the architecture to state-of-the-art approaches on a real-world data set.
Chronic Kidney Disease (CKD) has infected almost 800 million people around the world. Around 1.7 million people die each year because of it. Detecting CKD in the initial stage is essential for saving millions of lives. Many researchers have applied distinct Machine Learning (ML) methods to detect CKD at an early stage, but detailed studies are still missing. We present a structured and thorough method for dealing with the complexities of medical data with optimal performance. Besides, this study will assist researchers in producing clear ideas on the medical data preparation pipeline. In this paper, we applied KNN Imputation to impute missing values, Local Outlier Factor to remove outliers, SMOTE to handle data imbalance, K-stratified K-fold Cross-validation to validate the ML models, and a novel hybrid feature selection method to remove redundant features. Applied algorithms in this study are Support Vector Machine, Gaussian Naive Bayes, Decision Tree, Random Forest, Logistic Regression, K-Nearest Neighbor, Gradient Boosting, Adaptive Boosting, and Extreme Gradient Boosting. Finally, the Random Forest can detect CKD with 100% accuracy without any data leakage.
Conditional variational autoencoders (CVAEs) are versatile deep generative models that extend the standard VAE framework by conditioning the generative model with auxiliary covariates. The original CVAE model assumes that the data samples are independent, whereas more recent conditional VAE models, such as the Gaussian process (GP) prior VAEs, can account for complex correlation structures across all data samples. While several methods have been proposed to learn standard VAEs from partially observed datasets, these methods fall short for conditional VAEs. In this work, we propose a method to learn conditional VAEs from datasets in which auxiliary covariates can contain missing values as well. The proposed method augments the conditional VAEs with a prior distribution for the missing covariates and estimates their posterior using amortised variational inference. At training time, our method marginalises the uncertainty associated with the missing covariates while simultaneously maximising the evidence lower bound. We develop computationally efficient methods to learn CVAEs and GP prior VAEs that are compatible with mini-batching. Our experiments on simulated datasets as well as on a clinical trial study show that the proposed method outperforms previous methods in learning conditional VAEs from non-temporal, temporal, and longitudinal datasets.