A ANALISIS REGRESI DATA PANEL ANTARA FAKTOR JENIS IKAN TERHADAP VOLUME PRODUKSI PERIKANAN (TON) DI TINGKAT NASIONAL PERIODE TAHUN 2010-2020

REGRESSION ANALYSIS OF PANEL DATA BETWEEN FISH SPECIES FACTORS TO FISHERY PRODUCTION VOLUME (TON) AT THE NATIONAL LEVEL FOR THE PERIOD 2010-2020

Abstract

Regression analysis of panel data is a combination of cross section and time series data. The use of panel data can explain two kinds of information, namely information between units and between times. Panel data regression is a method to determine the effect of independent variables on dependent variables using Ordinary least Square (OLS) regression analysis modeling method. The potential of fishery production can be seen from the production of each type of fish produced by the community either through the production of aquaculture, marine capture fisheries and Fisheries capture mainland public waters (PUD). Fishery production tends to fluctuate from year. Increased fishery production is strongly influenced by the production factors of each type of fish as well as determined by the improvement of fishing facilities such as the addition of the fleet, changes in fishing gear and motorization of fishermen as well as improving the quality of resources and expertise of fishermen. In 2020 fisheries production at the national level covering 34 provinces was contributed by aquaculture ( 41 types of fisheries), marine capture (114 types of Fisheries), and 73 pud capture (types of Fisheries). In panel data regression there are three model estimates, namely CEM, FEM and REM. The CEM method is a method that assumes that The intercept and slope in each subject and each time are the same, the FEM method assumes that The intercept is different between subjects and the slope is the same between subjects, while the REM method assumes that the residual variable has a relationship between time and between subjects. In this study will be analyzed from 30 predictor variables of fish species that have the highest production in 2020, and what types of businesses significantly affect the volume of fishery production at the national level in 2010-2020. The results of this study, the best panel data regression model is using Random Effect Model (REM) with individual effects with R2 value of .....0,5117. Where the variable type of fish X1, X2,....X30 is able to explain the volume of fisheries production at the national level ......51.17 percent. The panel data regression equation is: Y = 914.61+ 0.885*X1 + 1.120*X2 + 1.323*X5 + 1.351*X4 + 1.772*X10 + 2.177*X12 + 1.714*X14 + 2.192*X15 + 1.085*X16 + 1.993*X18 + 1.896*X19 + 1,486*X20 + 2,330*x22 + 1,767*X25 + 1,961*X27 + 2,137*X28 + 2,096*X29. Where catfish variables significantly affect the volume of fishery production, the model shows the elasticity for the type of catfish is 0.885, which means that if the production of catfish increased by 1 ton, it will add the volume of fishery production by 0.885 tons. From the model results obtained, the strongest fishery production volume by province is West Sumatra, and the highest Volume is in Aceh province in 2020 at 17,628 tons, the lowest fishery production volume is in North Kalimantan province in 2010-2014 has no production.