Regression analysis assists in the prediction of a dependent variable from the relationship between the independent and dependent variable. From the above analysis the factor that affect the price of diamond.
The most is carat and certification. This is because the regression analysis of the above two factors displays a highly volatile and a rapid variation in the price if the price is plotted in the y axis and carats and certification on the x axis. The analysis further illustrates the difficulty of predicting the price variation due to the non consistency in price movements i.e. the price moves up and down complicating the prediction of the trend (John, 2002).
On the other hand factors like color, symmetry and polish have insignificant variation on the price of diamond. The price of diamond tends to be stable from the given that the graphical representation of price against these variables seem to be showing a horizontal linear variation. Since there are no huge variation in the prices given these factors, little emphasis and attention are given by the buyers while purchasing diamond.
The price of the given diamond is fair. Though the price is above the price as reflected I the regression analysis, the above variation is justified as it's not only the diamond but even the modeling of the ring is included. The price is further justified by the factors e.g. carats and clarity levels. Since the carats level is above 1 and the clarity is not the lowest, it is resolute for the price to be set at 3100 instead of the predicted value. Other factors as the color polish and certification also makes the price fair. This factor makes the prices to be fairly constant and reduces the high changes in prices. This is illustrated by the fairly horizontal and inelastic curve of price against this price.
The regression of color, carats and clarity against prices at the lower price levels are not significant to the 0.01 level of significance. This shows that the data is within the desired limit and therefore the regression is a model that appropriately helps in the prediction of the price of diamond. What causes the regression significance to be significant is the fact that the color and clarity are close determinants of the price while color is not a major factor for the determination of the diamonds price.
Hershey (1940) suggest that the entire run of clarity, color, certification and carats with prices shows fluctuating regression that continues to reflect the price changes in consideration of the factors. This run is best in the prediction of the prices as all the independent variables influencing the prices have been considered and the weight of each variable integrated in the analysis. This therefore gives a fair determinatio0n of prices compared top either regression analysis with few independent variables.
It is also worth noting that the regression analysis of carats with prices cannot be used in determination of prices when the carats level is 5.this is due to the fact that the 5 carats level used in the regression model is far below 5 and therefore other variables are also likely to vary significantly with the carats level of 5.ii is not possible to predict the price with the regression (Hesse, 2007. pp 42).
Apart from assigning numbers to the clarity levels, we could also consider reducing the number of rankings by coalescing close rankings and doing the rankings as one. This would make the regression simpler but there is hard to determine the criteria of merging the levels. Again, consideration could be given to the two extreme levels of clarity rather than all the levels. It suffers the drawback of not considering all information. The method has the advantage of being faster to use and determine extremes.