Predictive modeling: least squares method for compression of time-series data

1997 ◽  
Author(s):  
Saraswathi Mukherjee ◽  
Justin Zobel
Keyword(s):  
Time Series ◽  
Least Squares ◽  
Predictive Modeling ◽  
Time Series Data ◽  
Least Squares Method ◽  
Series Data

scholarly journals Associations between Meteorological Factors and Reported Mumps Cases from 1999 to 2020 in Japan

Epidemiologia ◽  
2021 ◽  
Vol 2 (2) ◽  
pp. 162-178
Author(s):  
Keiji Mise ◽  
Ayako Sumi ◽  
Shintaro Takatsuka ◽  
Shin-ichi Toyoda
Keyword(s):  
Time Series ◽  
Least Squares ◽  
Time Series Data ◽  
Least Squares Method ◽  
Meteorological Factors ◽  
Vaccine Coverage ◽  
Meteorological Data ◽  
Spectral Lines ◽  
Series Data ◽  
Mean Temperature

The present study investigated associations between epidemiological mumps patterns and meteorological factors in Japan. We used mumps surveillance data and meteorological data from all 47 prefectures of Japan from 1999 to 2020. A time-series analysis incorporating spectral analysis and the least-squares method was adopted. In all power spectral densities for the 47 prefectures, spectral lines were observed at frequency positions corresponding to 1-year and 6-month cycles. Optimum least-squares fitting (LSF) curves calculated with the 1-year and 6-month cycles explained the underlying variation in the mumps data. The LSF curves reproduced bimodal and unimodal cycles that are clearly observed in northern and southern Japan, respectively. In investigating factors associated with the seasonality of mumps epidemics, we defined the contribution ratios of a 1-year cycle (Q1) and 6-month cycle (Q2) as the contributions of amplitudes of 1-year and 6-month cycles, respectively, to the entire amplitude of the time series data. Q1 and Q2 were significantly correlated with annual mean temperature. The vaccine coverage rate of a measles–mumps–rubella vaccine might not have affected the 1-year and 6-month modes of the time series data. The results of the study suggest an association between mean temperature and mumps epidemics in Japan.

Predictive modeling for vehicle time-series data

2016 ◽  
pp. 477-482
Author(s):  
J. Wallner ◽  
F. Diermeyer ◽  
S. Engel
Keyword(s):  
Time Series ◽  
Predictive Modeling ◽  
Time Series Data ◽  
Series Data

scholarly journals A Novel Time Series Prediction Approach Based on a Hybridization of Least Squares Support Vector Regression and Swarm Intelligence

2014 ◽  
Vol 2014 ◽  
pp. 1-8 ◽  
Author(s):  
Nhat-Duc Hoang ◽  
Anh-Duc Pham ◽  
Minh-Tu Cao
Keyword(s):  
Time Series ◽  
Least Squares ◽  
Support Vector Regression ◽  
Time Series Data ◽  
Time Series Prediction ◽  
Hybrid Approach ◽  
Series Data ◽  
Support Vector ◽  
Promising Alternative ◽  
Prediction Approach

This research aims at establishing a novel hybrid artificial intelligence (AI) approach, named as firefly-tuned least squares support vector regression for time series prediction(FLSVRTSP). The proposed model utilizes the least squares support vector regression (LS-SVR) as a supervised learning technique to generalize the mapping function between input and output of time series data. In order to optimize the LS-SVR’s tuning parameters, theFLSVRTSPincorporates the firefly algorithm (FA) as the search engine. Consequently, the newly construction model can learn from historical data and carry out prediction autonomously without any prior knowledge in parameter setting. Experimental results and comparison have demonstrated that theFLSVRTSPhas achieved a significant improvement in forecasting accuracy when predicting both artificial and real-world time series data. Hence, the proposed hybrid approach is a promising alternative for assisting decision-makers to better cope with time series prediction.

scholarly journals Estimating the SARS-CoV-2 infected population fraction and the infection to-fatality ratio: A data-driven case study based on Swedish time series data

2021 ◽  
Author(s):  
Andreas Wacker ◽  
Anna Jöud ◽  
Bo Bernhardsson ◽  
Philip Gerlee ◽  
Fredrik Gustafsson ◽  
...  
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Least Squares Method ◽  
Finite Impulse Response ◽  
Herd Immunity ◽  
Late Summer ◽  
Epidemiological Data ◽  
Series Data ◽  
Filter Function ◽  
Study Results

Aim: To estimate the COVID-19 infection-to-fatality ratio (IFR), infection-to-case ratio (ICR), and infection-to-ICU admission ratio (IIAR) in Sweden; to suggest methods for time series reconstruction and prediction. Methods: We optimize a set of simple finite impulse response (FIR) models comprising of a scaling factor and time-delay between officially reported cases, ICU admissions and deaths time series using the least squares method. Combined with randomized PCR study results, we utilize this simple model to estimate the total number of infections in Sweden, and the corresponding IFR. Results: The model class provides a good fit between ICU admissions and deaths throughout 2020. Cases fit consistently from July 2020, by when PCR tests had become broadly available. We observe a diminished IFR in late summer as well as a strong decline during 2021, following the launch of a nation-wide vaccination program. The total number of infections during 2020 is estimated to $1.3$ million. Conclusions: A FIR model with a delta filter function describes the evolution of epidemiological data in Sweden well. The fact that we found IFR, ICR and IIAR constant over large parts of 2020 is in contrast with claims of healthcare adaptation or mutated virus variants importantly affecting these ratios. The model allows us to retrospectively estimate the COVID-19 epidemiological trajectory, and conclude that Sweden was far from herd immunity by the end of 2020.

scholarly journals Determinants of Pro-Poor Growth and Its Impacts on Income Share: Evidence from Ethiopian Time Series Data

2021 ◽  
Vol 2021 ◽  
pp. 1-7
Author(s):  
Gemechu Bekana Fufa
Keyword(s):  
Time Series ◽  
Human Capital ◽  
Time Series Data ◽  
Least Squares Method ◽  
Income Group ◽  
Ordinary Least Squares ◽  
Series Data ◽  
Poor Growth ◽  
Negative Impacts ◽  
Positive Impacts

The growing research interest in the pro-poorness of growth is the main issue today. Reducing economic poverty and inequality through pro-poor growth is the aim of policies in many countries. Pro-poor growth is good for poverty eradication if it can be achieved. Ethiopia is a good example of a country where growth was pro-poor between 1990 and 2018 but the pro-poor growth was reversed in 2016. The paper examined what led to pro-poor growth between 1990 and 2018 and what may have been responsible for the reversal in 2016. Unit root test reveals that all the series are nonstationary at level and stationary at first difference and have one cointegration relation between the variables. The dynamic ordinary least squares method was used to analyze the Ethiopian time series data from World Bank Development Indicators between 1990 and 2018 for the determinant of pro-poor growth. Regression analysis shows that job creation was responsible for the pro-poor growth between 1990 and 2018. The results of the analysis showed that human capital, industrial, and services growth have negative impacts on poorest people, whereas employment and agriculture growth have positive impacts on poorest people. In the richest income group, human capital, and industrial and service growths have positive impacts while agricultural growth and employment have negative impacts.

Time Series Regression with Mixtures of Integrated Processes

1995 ◽  
Vol 11 (5) ◽  
pp. 1033-1094 ◽  
Author(s):  
Yoosoon Chang ◽  
Peter C.B. Phillips
Keyword(s):  
Time Series ◽  
Least Squares ◽  
Time Series Data ◽  
Unit Roots ◽  
Ordinary Least Squares ◽  
Series Data ◽  
Vector Autoregressions ◽  
Time Series Regression ◽  
First Order ◽  
Unknown Mixture

The paper develops a statistical theory for regressions with integrated regressors of unknown order and unknown cointegrating dimension. In practice, we are often unsure whether unit roots or cointegration is present in time series data, and we are also uncertain about the order of integration in some cases. This paper addresses issues of estimation and inference in cases of such uncertainty. Phillips (1995, Econometrica 63, 1023–1078) developed a theory for time series regressions with an unknown mixture of 1(0) and 1(1) variables and established that the method of fully modified ordinary least squares (FM-OLS) is applicable to models (including vector autoregressions) with some unit roots and unknown cointegrating rank. This paper extends these results to models that contain some I(0), I(1), and I(2) regressors. The theory and methods here are applicable to cointegrating regressions that include unknown numbers of I(0), I(1), and I(2) variables and an unknown degree of cointegration. Such models require a somewhat different approach than that of Phillips (1995). The paper proposes a residual-based fully modified ordinary least-squares (RBFMOLS) procedure, which employs residuals from a first-order autoregression of the first differences of the entire regressor set in the construction of the FMOLS estimator. The asymptotic theory for the RBFM-OLS estimator is developed and is shown to be normal for all the stationary coefficients and mixed normal for all the nonstationary coefficients. Under Gaussian assumptions, estimation of the cointegration space by RBFM-OLS is optimal even though the dimension of the space is unknown.

Sign in / Sign up

Export Citation Format

Share Document