An Analysis on the Impact of Natural Disasters on the Economy of the Philippines

Over the past century, natural disasters have been terrorizing the economy by causing human fatalities and damaging infrastructure and production inputs. The Solow growth model suggests that natural disasters adversely affect gross domestic product (GDP) since these disrupt the production of inputs. On the contrary, the Schumpeterian growth theory provides an explanation behind the positive effect of natural disasters on economic growth. This study analyzed the relationship between natural disasters (i.e. earthquake, flood, and storm), economic activities (i.e. foreign aid and foreign direct investment) and GDP per capita income in the Philippines from 1990 to 2019. This study employed a multivariate analysis, time series regression, and autoregressive distributed lag (ARDL) approach. The results revealed a complex relationship between GDP per capita and the regressors. In the short run, the independent variables have a negative and significant relationship with the country’s per capita income. On the contrary, only FDI has a significant long-run relationship with the economy of the Philippines. The results highlight the Philippines’ need for comprehensive disaster plans and to lessen its dependence on foreign and external factors.

Forecasting Stock Price PT. Telkom Using Hybrid Time Series Regression Linear– Autoregressive Integrated Moving Average Model

The hybrid method is a method of combining two forecasting models. Hybrid method is used to improve forecasting accuracy. In this study, the Time Series Regression (TSR) linear model will be combined with the Autoregressive Integrated Moving Average (ARIMA) model. The TSR linear model is used to obtain the model and residual value, then the residual value of the TSR linear model will be modeled by the ARIMA model. This combination method will produce a hybrid TSR linear-ARIMA model. The case study in this research is stock closing price (daily) of PT. Telkom Indonesia Tbk. The stock closing price (daily) of PT. Telkom Indonesia Tbk in 2020 showed an decreasing and increasing trend pattern. The results of this study obtained the best model of hybrid TSR linear-ARIMA (2,1,1) with the proportion of data training and testing is 70:30. In the best model, the MAD value is 56.595, the MAPE value is 1.880%, and the RMSE value is 78.663. It is also found that the hybrid TSR linear-ARIMA model has a smaller error value than the TSR linear model. The results of forecasting the stock price of PT. Telkom Indonesia Tbk for the period 02 January 2021 to 29 January 2021 formed a decreasing trend pattern.

Application of the Autoregressive Integrated Moving Average Exogenous (ARIMAX) with Calendar Variation Effect Method for Forecasting Chocolate Data in Indonesia and the United States

Forecasting chocolate consumption is required by producers in preparing the amount of production each month. The tradition of Valentine, Christmas and Eid al-Fitr which are closely related to chocolate makes it impossible to predict chocolate by using the Classical Time Series method. Especially for Eid al-Fitr, the determination follows the Hijri calendar and each year advances 10 days on the Masehi calendar, so that every three years Eid al-Fitr will occur in a different month. Based on this, the chocolate forecasting will show a variation calendar effect. The method used in modeling and forecasting chocolate in Indonesia and the United States is the ARIMAX (Autoregressive Integrated Moving Average Exogenous) method with Calendar Variation effect. As a comparison, modeling and forecasting are also carried out using the Naïve Trend Linear, Naïve Trend Exponential, Double Exponential Smoothing, Time Series Regression, and ARIMA methods. The ARIMAX method with Calendar Variation Effect produces a very precise MAPE value in predicting chocolate data in Indonesia and the United States. The resulting MAPE value is below 10 percent, so it can be concluded that this method has a very good ability in forecasting.

Forecasting the Existence of Chocolate with Variation and Seasonal Calendar Effects Using the Classic Time Series Approach

Chocolate is the raw material for making cakes, so consumption of chocolate also increases on Eid al-Fitr. However, this is different in the United States where the tradition of sharing chocolate cake is carried out on Christmas. To monitor the existence of this chocolate can be through the movement of data on Google Trends. This study aims to predict the existence of chocolate from the Google trend where the use of chocolate by the community fluctuates according to the calendar variance and seasonal rhythm. The method used is classic time series, namely nave, double exponential smoothing, multiplicative decomposition, addictive decomposition, holt winter multiplicative, holt winter addictive, time series regression, hybrid time series, ARIMA, and ARIMAX. Based on MAPE in sample, the best time series model to model the existence of chocolate in Indonesia is ARIMAX (1,0,0) while for the United States it is Hybrid Time Series Regression-ARIMA(2,1,[10]). For forecasting the existence of chocolate in Indonesia, the best models in forecasting are ARIMA (([11],[12]),1,1) and Naïve Seasonal. In contrast to the best forecasting model for the existence of chocolate in the United States, namely Hybrid Naïve Seasonal-SARIMA (2,1,0)(0,0,1)12 Hybrid Time Series Regression- ARIMA(2,1,[10]), Time Series Regression, Winter Multiplicative, ARIMAX([3],0,0).

Climate Finance: Mapping Air Pollution and Finance Market in Time Series

Climate finance is growing popular in addressing challenges of climate change because it controls the funding and resources to emission entities and promotes green manufacturing. In this study, we determined that PM2.5, PM10, SO2, NO2, CO, and O3 are the target pollutant in the atmosphere and we use a deep neural network to enhance the regression analysis in order to investigate the relationship between air pollution and stock prices of the targeted manufacturer. We also conduct time series analysis based on air pollution and heavy industry manufacturing in China, as the country is facing serious air pollution problems. Our study uses Convolutional-Long Short Term Memory in 2 Dimension (ConvLSTM2D) to extract the features from air pollution and enhance the time series regression in the financial market. The main contribution in our paper is discovering a feature term that impacts the stock price in the financial market, particularly for the companies that are highly impacted by the local environment. We offer a higher accurate model than the traditional time series in the stock price prediction by considering the environmental factor. The experimental results suggest that there is a negative linear relationship between air pollution and the stock market, which demonstrates that air pollution has a negative effect on the financial market. It promotes the manufacturer’s improving their emission recycling and encourages them to invest in green manufacture—otherwise, the drop in stock price will impact the company funding process.

Impact of an electronic alert on prescription patterns of meropenem, voriconazole and caspofungin

Background Antimicrobial stewardship programs promote the appropriate use of antimicrobial substances through the implementation of evidence-based, active and passive interventions. We analyzed the effect of a computer-assisted intervention on antimicrobial use in a tertiary care hospital. Methods Between 2011 and 2016 we introduced an electronic alert for patients being prescribed meropenem, voriconazole and caspofungin. At prescription and at day 3 of treatment, physicians were informed about the risk related to these antimicrobial substances by an electronic alert in the medical records. Physicians were invited to revoke or confirm the prescription and to contact the infectious disease (ID) team. Using interrupted time series regression, the days of therapy (DOTs) and the number of prescriptions before and after the intervention were compared. Results We counted 64,281 DOTs for 5549 prescriptions during 4100 hospital stays. Overall, the DOTs decreased continuously over time. An additional benefit of the alert could not be observed. Similarly, the number of prescriptions decreased over time, without significant effect of the intervention. When considering the three drugs separately, the alert impacted the duration (change in slope of DOTs/1000 bed days; P = 0.0017) as well as the number of prescriptions (change in slope of prescriptions/1000 bed days; P < 0.001) of voriconazole only. Conclusions The introduction of the alert lowered prescriptions of voriconazole only. Thus, self-stewardship alone seems to have a limited impact on electronic prescriptions of anti-infective substances. Additional measures such as face-to-face prompting with ID physicians or audit and feedback are indispensable to optimize antimicrobial use.

Model Variasi Kalender pada Regresi Runtun Waktu untuk Peramalan Jumlah Pengunjung Grojogan Sewu

<p><strong>Abstract.</strong> Grojogan Sewu visitors experience a significant increase during school holidays, year-end holidays, and also Eid al-Fitr holidays. The determination of Eid Al-Fitr uses the Hijriyah calendar so that the occurrence of Eid al-Fitr will progress 10 days when viewed from the Gregorian calendar, this causes calendar variations. The objective of this paper is to apply a calendar variation model based on time series regression and SARIMA models for forecasting the number of visitors in Grojogan Sewu. The data are Grojogan Sewu visitors from January 2009 until December 2019. The results show that time series regression with calendar variation yields a better forecast compared to the SARIMA model. It can be seen from the value of root mean square error (<em>RMSE</em>) out-sample of time series regression with calendar variation is less than of SARIMA model.</p><p><strong>Keywords: </strong>Calendar variation, time series regression, SARIMA, Grojogan Sewu</p>

Negative Binomial Time Series Regression – Random Forest Ensemble in Intermittent Data

Intermittent dataset is a unique data that will be challenging to forecast. Because the data is containing a lot of zeros. The kind of intermittent data can be sales data and rainfall data. Because both sometimes no data recorded in a certain period. In this research, the model is created to overcome the problem. The approach that is used in this research is the ensemble method. Mostly the intermittent data comes from the Negative Binomial because the variance is over the mean. We use two datasets, which are rainfall and sales data. So, our approach is creating the base model from the time series regression with Negative Binomial based, and then we augmented the base model with a tree-based model which is random forest. Furthermore, we compare the result with the benchmark method which is The Croston method and Single Exponential Smoothing (SES). As the result, our approach can overcome the benchmark based on metric value by 1.79 and 7.18.