Imputation for the analysis of missing values and prediction of time series data

Sridevi, S.; Rajaram, S.; Parthiban, Chembian; SibiArasan, S.; Swadhikar, C.

doi:10.1109/icrtit.2011.5972466

Cited by 22 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the following experiments, we choose the closest observation to the missing value as the candidate. AR imputation is proposed in [15] and it uses auto-regression model (p = 4) to fill in missing values. They are tested on 3 generated datasets (sine function, sinc function, and Mackey-Glass chaotic time series) and 4 benchmark datasets (Poland electricity load, Sunspot, Jenkins-Box, and EUNITE competition).…”

Section: Resultsmentioning

confidence: 99%

Time Series Forecasting with Missing Values

Wu¹,

Chang²,

Lee³

2015

Proceedings of the 1st International Conference on Industrial Networks and Intelligent Systems

View full text Add to dashboard Cite

Abstract-Time series prediction has become more popular in various kinds of applications such as weather prediction, control engineering, financial analysis, industrial monitoring, etc. To deal with real-world problems, we are often faced with missing values in the data due to sensor malfunctions or human errors. Traditionally, the missing values are simply omitted or replaced by means of imputation methods. However, omitting those missing values may cause temporal discontinuity. Imputation methods, on the other hand, may alter the original time series. In this study, we propose a novel forecasting method based on least squares support vector machine (LSSVM). We employ the input patterns with the temporal information which is defined as local time index (LTI). Time series data as well as local time indexes are fed to LSSVM for doing forecasting without imputation. We compare the forecasting performance of our method with other imputation methods. Experimental results show that the proposed method is promising and is worth further investigations.

show abstract

Section: Resultsmentioning

confidence: 99%

Time Series Forecasting with Missing Values

Wu¹,

Chang²,

Lee³

2015

Proceedings of the 1st International Conference on Industrial Networks and Intelligent Systems

View full text Add to dashboard Cite

show abstract

“…The work [4] focuses on the development of ARLSimpute which represents the autoregressive model for predicting missing values. The output of the data preprocessing is provided as input to the prediction techniques namely quadratic & linear prediction.…”

Section: Related Workmentioning

confidence: 99%

Regression Based Data Pre-processing Technique for Predicting Missing Values

Shastry¹,

Sanjay²,

Praveen

2021

Emerging Research in Computing, Information, Communication and Applications

View full text Add to dashboard Cite

“…Mean Imputation fills in missing data by using the mean value of the observed values of the variable. This method affects the statistical characteristics of the dataset, specifically by reducing the variance and the standard deviation [14]. On the one hand, mean imputation results in a dataset of the same size after the imputation and the same mean as the original dataset, but on the other hand, it results in an underestimation of the standard deviation.…”

Section: Mean Imputationmentioning

confidence: 99%

“…14 gives an example of Multiple Imputation where m = 2, which generates two complete datasets.Figure 2.15 shows the single complete dataset generated after single imputation.…”

mentioning

confidence: 99%