-
Method of Disaggregating Annual Time Series into Seasons
Richard Hindls,
Stanislava Hronova
Issue:
Volume 9, Issue 1, March 2023
Pages:
1-8
Received:
15 January 2023
Accepted:
1 February 2023
Published:
9 February 2023
Abstract: A number of indicators (especially economic ones), whose values are monitored with annual periodicity, need to be broken down into the so-called seasons, i.e., time periods shorter than one year. And it is necessary not only to respect the specifics of these seasons, but at the same time to do so without having to specifically survey these values in the seasons. The aim of the paper is to propose a method for distributing the annual values of an indicator into seasons (quarters, months) using a suitably chosen indicator, without subsequent correction. This will preserve the link to the processes evolving within the year, while removing the formalism of splitting the residual generated in the first step. The paper presents a new approach, in which simulated values of the response variables enable us to estimate the series' parameters with the aid of a simple loss function. Such a solution reduces the dependence of formal models on real data, which may or may not be available to the official statisticians who are responsible for statistical surveys. The paper shows the theoretical concept of the disaggregation method, which is the result of research in the given area and was verified on the example of data for the Czech Republic.
Abstract: A number of indicators (especially economic ones), whose values are monitored with annual periodicity, need to be broken down into the so-called seasons, i.e., time periods shorter than one year. And it is necessary not only to respect the specifics of these seasons, but at the same time to do so without having to specifically survey these values i...
Show More
-
Bayesian Multiple Linear Regression Model for GDP in Nepal
Ranjita Pandey,
Dipendra Bahadur Chand,
Himanshu Tolani
Issue:
Volume 9, Issue 1, March 2023
Pages:
9-23
Received:
13 January 2023
Accepted:
6 February 2023
Published:
27 February 2023
Abstract: Gross Domestic Product (GDP) known as the pulse of economy for any country depends on multiple factors like export-import, inflation rate and unemployment rate etc. Statistical assessment of GDP demands fresh concepts to explain GDP through its covariates in order to improve and strengthen estimation process. Descriptive analysis for the considered data set from world bank for GDP and covariates is presented through Heatmaps. Identification and relevance of possible set of covariates is done by Ordinary Least Square (OLS) regression followed by Step-wise regression. We propose an alternative statistical algorithm implemented as Bayesian Inference through Integrated Nested Laplace Approximation (INLA) which bridges the gap of accuracy in estimates as opposed to frequentist OLS regression for explaining GDP of country Nepal. Effect of changing prior parameters is assessed through Deviance Information Criterion (DIC). Different scenarios for prior distribution for regression parameters were analyzed to identify most suitable choice of parameter for normal distribution. The comparison of Bayesian and frequentist modelling results is done using several criteria such as Mean Square Error (MSE), Mean Absolute Deviation (MAD) and Mean Absolute Percent Error (MAPE). Bayesian estimation approach is a more efficient method for parametric estimation as compared to OLS classical method. GDP of Nepal was found to have strongest relationship with unemployment rate of Nepal as evident from both classical and Bayesian model.
Abstract: Gross Domestic Product (GDP) known as the pulse of economy for any country depends on multiple factors like export-import, inflation rate and unemployment rate etc. Statistical assessment of GDP demands fresh concepts to explain GDP through its covariates in order to improve and strengthen estimation process. Descriptive analysis for the considered...
Show More
-
Analyzing Dynamic Regimes of GARCH Model on Stock Price Volatility
Rosemary Ukamaka Okafor,
Josephine Nneamaka Onyeka-Ubaka
Issue:
Volume 9, Issue 1, March 2023
Pages:
24-34
Received:
20 January 2023
Accepted:
16 February 2023
Published:
27 February 2023
Abstract: As a result of volatility dynamics, investors and other stakeholders in businesses and industries have difficulty making financial decisions. Generalized Autoregressive Conditional Heteroskedasticity (GARCH) models are the most widely applied in the analysis of financial derivatives volatility. Volatility persistence is a common issue when analyzing stock prices, making it cumbersome for GARCH models. The GARCH model is transformed into the Makov switching GARCH model to check for dynamics in volatility persistence. Markov Regime-Switching GARCH (MSGARCH) models permit the conditional mean and variance to change across regimes over time. The Markov switching GARCH models incorporate the regime variables in the parameter space, making it viable for the parameters to be estimated by the maximum likelihood estimation method, unlike the classical GARCH models. Zenith Bank plc’s daily closing stock prices, a top-tier stock on the Nigerian Stock Exchange market, are fitted using the GARCH and MSGARCH models. The comparison between the MSGARCH model and the classical GARCH model was verified using the AIC and BIC metrics as well as the one with the maximum log likelihood estimates. The outcome suggests that MSGARCH model performs better than the single-regime GARCH model and that it yields significantly better out of-sample volatility forecasts. The results will aid the stakeholders to leverage and mitigate risks in their investment on the selected stocks.
Abstract: As a result of volatility dynamics, investors and other stakeholders in businesses and industries have difficulty making financial decisions. Generalized Autoregressive Conditional Heteroskedasticity (GARCH) models are the most widely applied in the analysis of financial derivatives volatility. Volatility persistence is a common issue when analyzin...
Show More
-
Two Sample Approaches to Regression Calibration for Measurement Error Correction
Samuel Joel Kamun,
Cornelious Nyakundi,
Richard Simwa
Issue:
Volume 9, Issue 1, March 2023
Pages:
35-40
Received:
1 February 2023
Accepted:
1 March 2023
Published:
9 March 2023
Abstract: The goal of this work is to create methods for enhancing measurement error using regression calibration as a strategy by combining two samples, thereby increasing the relative efficiency of linear regression models. Because two or more samples are more likely to provide an accurate representation of the population than a single sample under inquiry, utilizing two samples in regression calibration is likely to produce a realistic depiction of what the actual population is when error-free. This study has generated independent estimates from two samples and combined them with weights equal to the inverse of their estimated probabilities of sample inclusion. It has also integrated two data sets into a single data set and suitably adjusted the weights on each sampled unit. The regression calibration method is most commonly used to correct predictor-response bias caused by variable measurement imperfections. Because of its simplicity, this method is often used. The fundamental principle behind regression calibration is to estimate the conditional expectation of a genuine response, given predictors measured with error and other covariates supposed to be measured without error. The predicted values are then estimated and used to assess the relationship between the response and an outcome in place of the unknown genuine response. Further information on the unobservable true predictors is required by the regression calibration program. This data is frequently obtained from a validation study that employs unbiased measurements for genuine predictors. This study has employed and compared the results obtained from the two sample approaches. Measuring errors can be produced by a variety of sources, including instrument error, laboratory error, human error, problems in documenting or executing measurements, self-reporting errors, and natural oscillations in the underlying amount. Covariate measurement error has three effects: In addition to hiding the properties of the data, which makes graphical model analysis difficult, it produces bias in parameter estimates for statistical models, resulting in a sometimes significant loss of power for detecting fascinating correlations between variables. The two sample approaches employed by the study have yielded acceptable results.
Abstract: The goal of this work is to create methods for enhancing measurement error using regression calibration as a strategy by combining two samples, thereby increasing the relative efficiency of linear regression models. Because two or more samples are more likely to provide an accurate representation of the population than a single sample under inquiry...
Show More
-
Bayesian Analysis on the Spatial Difference of Input Risk of Overseas Cases of COVID-19 in China
Bo Yang,
Yunyuan Yang,
Wei Zheng,
Yanmei Li,
Xinping Yang
Issue:
Volume 9, Issue 1, March 2023
Pages:
41-48
Received:
24 February 2023
Accepted:
20 March 2023
Published:
31 March 2023
Abstract: To analyze the spatial difference of COVID-19 import risk is helpful for scientific prevention and control. On the basis of clustering 25 provinces and cities with epidemic input in study time, a multinomial distribution model was established under the Bayesian framework. All parameters Bayesian estimation was obtained by MCMC method. 25 provinces and cities with overseas input were divided into 9 categories from March 3 to April 23, 2020. 468 overseas input risk values are regarded as parameters, and the maximum MC-error estimated by Bayesian is only 0.677% of the standard deviation. During the study period, 25 provinces and cities have input risk. The highest risk areas of overseas import are 12 provinces and cities in the first category represented by Beijing, Shanghai and Guangdong Province, including 10 provinces and cities along the coast / border. The lowest risk areas are the eighth category (Henan Province) and the ninth category (Anhui Province); the fourth category (Heilongjiang Province and Shanxi Province) risk is higher than the first category in 7 days and it has the largest input vary fluctuation. Taking 2020-3-22, 4-7 and 4-18 as time nodes, the overseas input risk is divided into four stages. In the first stages, the highest risk of overseas import is the first category (59.613%); in the second and third stages are the first category (decline from 60.505% to 37.056%), the fourth category (increase from 16.071% to 33.852%); in the fourth stage, the first category (42.622%), the third category (Shaanxi Province and Jilin Province, 17.556%) and the fourth category (10.056%).
Abstract: To analyze the spatial difference of COVID-19 import risk is helpful for scientific prevention and control. On the basis of clustering 25 provinces and cities with epidemic input in study time, a multinomial distribution model was established under the Bayesian framework. All parameters Bayesian estimation was obtained by MCMC method. 25 provinces ...
Show More