Based on the downloaded econometric working paper, entitled: The Econometric Analysis of Factors Affecting the Demand for Frozen Food, the causal relationship will exist when a dependent variable gives effects to the independent variables. In this paper, the author did the analysis on causal relationship between the factors affecting the demand for frozen food, such as education level of the mother, number of family members, status of the mother, possession of a house, possession of an automobile, possession of deep freeze, life period of the family and the income level. From the analysis, the cause variable is the factors affecting the demand for frozen food while the effect variable is the demand for frozen food.

From the paper, the correlation between the cause and effect variable varies, which is positive and negative correlation. There is positive correlation between the variable mother’s level of education, possession of a house and an automobile with the demand for the frozen food. This can be proven because, the more educated the mother, when an individual own a house and mobile, there will be higher demand for them towards the frozen food. Moreover, the income level shows the positive correlation towards the effect variable, when someone’s income increase, the demand will be increase too. For the variable status of the mother, a married mother will demand more because of her family. However, number of family members, possession of deep freeze and life period of the family were negatively correlated with the effect variable. But we cannot say that these variables show negative effect, there must be other effects that might control the dependent variable. Negatively correlation does not mean that the variables are not important at all, it just like they do not give too much effect to the model.

The type of data used is the econometric model in this working paper is cross sectional data, because ‘simple random sampling’ method used in this model. Besides, cross sectional data consists of a sample of unit, taken at a given point in time. Random sampling method is one of the significant features that obtained from a cross sectional data. For this case of analysis, the information was obtained from 336 people from the population by random sampling.

In my opinion, the zero mean conditional assumption applies to this econometric model in the working paper. This is because despite of all the factors mentioned, there are few other factors that left in u, unobserved factor. Unfortunately, we never know whether the value or variable in u is unrelated to the explanatory variables. Furthermore, based on a journal entitled American Journal of Food Science and Nutrition – ( http://www.openscienceonline.com/journal/fsnr) , the other factors that might give effects to the demand for the frozen food are the price of the frozen food, marital status of an individual and the distance from their place to the market.

From the result in this econometric model, the signs and magnitudes of the estimated coefficient in the explanatory variables vary. The coefficient on ELM, education level of the mother, 1.097 is in accordance with the statement that the higher the education level of the mother leads to a high demand for the frozen food. Furthermore, the coefficient on SM, status of mother, PH, possession of a house, PA, possession of an automobile and IL, income level also in positive sign which is 0.103, 0.624, 0.063 and 2.047. However, the coefficients on NFM, number of family members, PDF, possession of deep freeze and LPF, life period of the family is -0.527, -0.339, and -0.638 respectively. From this, we know that those explanatory variables are inversely proportional with the effect variable.

Since df = n – k – 1 = 366 – 8 – 1 = 357. The degree of freedom value is large, so the critical value must be find using the standard normal distribution. At, the 5% significance level, the critical value is 1.65. The test for statistically significant is as below: –

ELM, education level of mother

t-stat: 4.941 > 1.65

reject the null hypothesis that H_0: ?_2 = 0, in favour of H_A: ?_2 > 0

statistically significant at 5% significance level.

When the mother is an educated person, she knows well about the frozen food.

NFM, number of family members

t-stats: -3.046 0.

statistically significant at 5% significance level.

As number of family members increase, the consumption of frozen food will increase.

SM, status of the mother

t-stats: 0.1475 0.

statistically insignificant at 5% significance level.

Status of the mother would not be a big cause in the effect variable.

PH, possession of a house

t-stats: 1.592 0.

statistically insignificant at 5% significance level.

People that do not own a house should eat to live.

PA, possession of an automobile

t-stats: 0.089 0.

statistically insignificant at 5% significance level.

Clearly, we can see that those who do not possesses automobile will still have to consume foods.

PDF, possession of deep freeze

t-stats: -0.435 > -1.65

fail to reject the null hypothesis that H_0: ?_7 = 0 in favour of H_A: ?_7 > 0.

statistically insignificant at 5% significance level.

People can go out to buy the food and does not need to have a freeze.

LPF, life period of family

t-stats: -3.19 0.

statistically significant at 5% significance level.

Longer life period of the family leads to a high demand of the effect variable.

IL, income level of a family

t-stats: 5.130 > 1.65

reject the null hypothesis that H_0: ?_9 = 0 in favour of H_A: ?_9 > 0.

statistically significant at 5% significance level.

As income level increases, the demand for frozen food will increase too.

In my opinion, I think that the explanatory variables explain much variation in the dependent model. This is because the number of explanatory variables that are significant in the model is more than the insignificant one. Moreover, it has been stated in the paper that this model is valid and applicable.