Warta Informatika Pertanian : Volume 15, Desember 2006

 
  • Integrasi Pasar Kentang di Indonesia Analisis Korelasi dan Kointegrasi

Witono Adiyoga, Keith O Fuglie & Rahman Suherman, Balai Penelitian Tanaman Sayuran Lembang Bandung

Market integration in potato was examined by using correlation and co-integration approaches. Utilizing daily, weekly and monthly potato price data in Jakarta, Bandung, Tanah Karo and Singapore were analyzed to determine the extent and nature of market integration. In general, the study of market integration can provide information on how the markets operate, that may be useful for improving price policy, monitoring price movements, making price prediction, and improving infrastructure investment policy. Under current price situation, it was hypothesized that potato markets in Bandung, Jakarta, Tanah Karo and Singapore are not integrated. Results indicate that correlation coefficients are not unequivocal indicators of market integration.  High bi-variate correlation for two markets that do not trade with each other (segregated) is quite possible, if prices in each market are highly correlated via the price and trading relationship of a joint destination market. Further analysis suggests that the diagnostic approach of correlation coefficient in studying market integration should be used with cautious, because of some proven weaknesses that are inherent to this approach. The use of co-integration approach suggests that all of those markets are consistently integrated. The results are invariant to the type of data used (daily, weekly and monthly data). Co-integration is the statistical implication of the existence of a long-run relationship between economic variables (prices). In other words, from a statistical point of view, a long-term relationship means that the variables move together over time so that short-term disturbances from the long-term trend will be corrected. Integrated potato markets will help producers and consumers since the correct price signals can be transmitted down the supply chain. Consequently, consumers in some markets will not have to pay higher prices, and farmers will be able to grow a certain commodity according to their comparative advantages. This leads to more efficient use of resources.

Download pdf file (290 kb)
  • Pendugaan Produktivitas Tanaman Padi Sawah Melalui Analisis Citra Satelit 
Wahyunto, Widagdo & Bambang Heryanto, Balai Besar Litbang Sumberdaya Lahan Pertanian

Remote sensing technology is hopefully potential for monitoring rice growing stage, and rice yield estimation. The satellite remote sensing based on crop yield prediction is expected to attain greater importance as it provides information on smaller areas such as a district or sub district, before harvesting season which could not be achieved through the existing methods of crop yield estimation through ‘ubinan’ (crop cutting experiments=CCE). Landsat Thematic Mapper data were used for  detecting the spatial distribution  of rice area as these dates acquation (satellite over pass) represented the standing rice crops and measerues their greeness. In this study, a simple rice yield model was developed based on the relationship between estimated peak value of Normalized Difference Vegetation Index (NDVI) at the panicle formation (approx. 10-11 weeks after replanting) to the yield of several sample plots. Yield data of several sample plots were obtained by means of CCE taken from secondary data that were collected by Central Beureau Statistic and Agriculture Service. Then, NDVI statistics extracted from multidate (representing at the maximum greenness value) were correlated with yield data. Ordinary Least Square statistical approach was used to develope a rice yield estimation model. The estimated yield model regression statistics of  West Java and Central Java study areas are: yield (ton/ha)= 24,622x - 7,808 à R2=0,727 and yield (ton/ha)= 22,98x – 6,33 à R2=0,71 respectively, where x is NDVI value. Ground truth at sub-district (kecamatan) level were conducted to validate the rice yield estimation model. Yield estimation model were found to be promising in accuracy with deviation less than 5 percent or 0,27 ton from actual at Bekasi, Kerawang and Subang district, West Java and 6 percent or 0,31 ton at Demak district, Central Java. 

Download pdf file (329 kb)
  • Evaluasi Kinerja Manajemen Mendukung Perencanaan Penganggaran Program Litbang Pertanian
S. Asih Rohmani & Ketut G.M, Sekretariat Badan Litbang Jakarta

The objectives of evaluation of management performance are (1) to identify data to support research program management which will be used as a considerant in finance allocation in IAARD, (2) to implement an accountable research program within research institutes (3) to increase the appreciation toward performance variable of agricultural research program. Aspects to be used as variable of evaluation are planning, human resource development, capacity building, and research collaboration.  All variable were analized and a certain formula was applied in order to get a total score. The result of data analysis indicates that most of the research institutes (44,45%) have good performance, 12,69% of research institutes have an excellent performance and 42,86% research institutes are fair performance.

Download pdf file (231 kb)
  • Transformasi Data Skala Ordinal ke Interval dengan Menggunakan Makro Minitab
Budi Waryanto & Yuan Astika Millafati, Pusat Data dan Informasi Pertanian

Ordinal Data is frequently used in path analysis. In that case, the ordinal data is often  transformed to an interval data. The transformation process needs a convertion program, which is easy and fast. Makro minitab can be used in that transformation, however this program has a weakness. This article will discuss the improvement, which has been made in the tranformation of ordinal data to an interval data at least with data scale three (1-3, 1-5, 1-7, etc). In order to perform that task, a transformation program, namely “Konverum” is designed by using the makro Minitab. This program is able to transform ordinal data in Likert scale to an interval data, at least with scale 3 category. In addition, this program is also capable in transforming data wich is not completely filled out.

Download pdf file (603 kb)

  • Aplikasi Analisis Khi Kuadrat (X2) terhadap Kekurangan Energi Protein pada Anak di Bawah Lima Tahun (Balita) dan Faktor-Faktor yang Berhubungan 
M.E. Yusnandar dan Sri Sejati, Balai Penelitian Ternak Ciawi Bogor

Energy protein is essential nutrients for human growth which could be found in plant and animal. The chi square is one of statistical test tools which can be applied to do qualitative analyse or non parametric data by using cross-tab analysis method with two variables. The parameter which wants to be studied is lack of energy protein (LEP) of the children below five years old in relation to predisposition factors (education, knowledge and mother’s behaviour), proponent factor (infection of diseases), and reinforcement factor (family income and total family member). Two steps data analysis were done. First, frequencies distribution based on the list of the question being asked to the respondents and second step of data analysis, was categorizing of each variable. LEP (light and heavy) was set as dependent variable. In addition, education (low and high), knowledge (poor and good), mother behaviour (right and wrong), infection of diseases (infected and uninfected), family income (capable and un-capable) and total family member (large and small) were set as  independent variable. The results showed that relation of the LEP to the independent variables was significant (P<0.05). Its means that the LEP (Y) was influenced by education level (X1), knowledge (X2), mother behaviour (X3), infection of diseases (X4), family income (X5) and total family member (X6). The result of chi square were of LEP and Education : (Y–X1 ,  X20,05(r-1)(c-1) = 8,55. Based on this study, therefore, the factors which influence lack of energi for the children below 5 year old are the level of education, lack of knowledge, income and a member of household.

Download pdf file (423 kb)