ARIMA

In this case, the Gaussian Copula is used to connect the data that correlates with the time and with other data sets. Most often, practitioners rely only on the linear correlation to describe the degree of dependence between two or more variables; an approach that can lead to quite misleading conclusions as this measure is only capable of capturing linear relationships. Correlation doesn't mean causation, prediction using Copula is built on three things that the marginal distribution function, the kernel function, and the function of the Copula. Gaussian Copula involves the covariance matrix are approximated by using kernel functions. Kernel acts as the correlation between the approach of the data values that have the same characteristics. In this case, the characteristics used is the time. The advantage of the kernel function is able to calculate the correlation between random variables that have a realization using data characteristics. The advantage of using the kernel based Copula able to capture the dependencies between data and process data that have the same characteristics with time. Another benefit is that it allows a sequence of random variables have a joint distribution function so that the conditional probability of the prediction can be calculated. 1. PENDAHULUAN Karakteristik umum harga saham adalah memiliki tingkat ketidakpastian. Ketidakpastian ini dalam hal pergerakan harga saham dalam jangka pendek, ataupun dalam jangka panjang. Karakteristik ini tidak disukai oleh para investor karena menimbulkan risiko pada investasi mereka. Ketidakpastian ini juga tidak dapat dihindari dalam investasi. Untuk menghadapi ketidakpastian pergerakan harga, yang dapat dilakukan oleh investor adalah mengurangi ketidakpastian tersebut. Salah satu alat (tools) untuk mengurangi ketidakpastian adalah dengan melakukan prediksi (forecasting) harga saham. Proses prediksi harga saham terbagi menjadi dua kelompok. Kelompok pertama adalah pihak yang mempercayai bahwa terdapat suatu cara untuk memprediksi harga saham. Kelompok kedua adalah pihak yang berkeyakinan bahwa pasar adalah efisien, dan apabila ada informasi baru yang diperoleh, pasar akan menyerapnya dan mengkoreksi dirinya sendiri. Kelompok kedua ini percaya bahwa harga saham tidak dapat diprediksi. Hal inilah yang disebut teori Efficient Market Hypothesis (EMH) seperti yang disebutkan oleh Gryc (2006). Kelompok ini beranggapan bahwa pasar saham mengikuti pola random

Se utiliza la metodología CRISP de data mining aplicada sobre los datos abiertos gubernamentales de la COVID-19 para el caso de Perú y se emplean técnicas de series de tiempo para descubrir los mejores modelos que permitan realizar pronósticos respecto a los casos confirmados. Se aplican las fases de la metodología iterativamente: limpieza de datos, detección de hallazgos, selección del modelo Arima (autoregressive integrated moving average) para el análisis de series de tiempo y estimación de los parámetros que caracterizan la serie temporal. El estudio concluye que, para los datos abiertos gubernamentales de la COVID-19 en el caso de Perú, no es conveniente el uso de datos diarios porque existen diferencias significativas por el día de semana y que es mejor utilizar la semana. Sobre la base de ello, se determina que no existe un único modelo a nivel país ni a nivel departamento, por lo que se proponen modelos a nivel de provincias que resultan estadísticamente significativos para realizar pronósticos a corto plazo.

Historically, gold has been identified as a unique commodity because of its ability to act as a hedge against inflation and its stability during periods of financial volatility and crises. Tracing the trends in Gold prices and conducting a price forecast is relevant from the point of view of Gold producers, investors and various central banks in the world. Time series analysis is a popular forecasting method for predicting the future value of variables. Time series forecasting techniques attempt to account for changes in the variable over time by examining trends. This technique also uses historical data to predict the future value of a variable. The objective of this study was to develop a suitable forecasting model for the price of gold. ARIMA (Autoregressive Integrated Moving Average) model has been used in this paper for Gold price forecasting.

This study explored modelling and forecasting of wholesale groundnut monthly prices in Bikaner district of Rajasthan using Autoregressive Integrated Moving Average model (ARIMA) with and without external predictors, Seasonal Integrated Moving-Average Model (SARIMA) and Error Correction Models (ECM).Based on minimum AIC and BIC values, ARIMA (0, 1, 0) and SARIMA (2, 1, 1)*(0, 1, 1) were selected for forecasting. However, among the SARIMA models, SARIMA (0,1,1)*(0,1,1) had the lowest MAPE compared to SARIMA (1,1,1)*(1,1,1) and SARIMA (2, 1, 1)*(0, 1, 1) which had the lowest AIC and BIC values respectively. Also, ARIMA (2, 1, 2) showed the lowest MAPE of 7.095 which is lower than ARIMA (0, 1, 0) which had the lowest AIC and BIC, indicating lower AIC and BIC do not always give the best forecast. Furthermore, about 7.474 percent, 8.097 percent and 12.47 percent of forecasting error was shown by ARIMA (0,1,0), SARIMA (0,1,1)*(0,1,1) and ECM models respectively. Among the best fit selected models in this study, ARIMA showed higher potency in forecasting of wholesale groundnut prices in Bikaner. It is therefore suggested that a number of models should be used when forecasting prices of agricultural commodities and the best fit model selected. In addition, this finding has provided insight on how to forecast prices of agricultural commodities for the purposes of marketing intelligence which could be useful to farmers, wholesalers, retailers, consumers, government agencies and other stakeholders in groundnut trading especially in Rajasthan.