Anda di halaman 1dari 15

# Name of Institution

## Randomized Block Design

By: Amit Patwari Deepanshu Garg Nishant Arora Parteek Arora

What is RBD

Name of Institution

In the statistical theory of the design of experiments, blocking is the arranging of experimental units in groups (blocks) that are similar to one another. Data or Experiments have interrelation in some or the other way.

A completely randomized design is useful when the experimental units are homogenous.
If the experimental units are heterogeneous, blocking is often used to form homogeneous groups.

Name of Institution

## Process of Hypothesis Using RBD

1st step is Creating Null and Alternate Hypothesis, since F Test being used so Equality and Inequality will be there : Calculation of Matrices using Annova table.

The anova procedure for the randomized block design requires us to partition the sum of squares total (SST) into three groups i.e SST = SSTR + SSBL + SSE Where, SSTR = Sum of square due to treatments SSBL = Sum of square due to blocks SSE = Sum of square due to error Also the anova table shows how the -1 total degrees of freedom are apart such that k-1 degrees of freedom go to treatments,b-1 go to blocks . F = MSTR/ MSE

Name of Institution

Annova Table

Formulas Used

Name of Institution

Computations

Name of Institution

Name of Institution

The design is useful for comparing t treatment means in the presence of a single extraneous. Source of variability. It can, with eective grouping, provide substantially more precise results. Than a completely randomized design of comparable size.

## The statistical analysis is simple.

The design is easy to construct. It can be used to accommodate any number of treatments in any number of blocks

Name of Institution

Because the experimental units within a block must be homogeneous, the design is best suited. For a relatively small number of treatments. This design controls for only one extraneous source of variability (due to blocks). Additional extraneous sources of variability trend to increase the error term, making it more dicult to detect treatment dierences.

## Data and Its Analysis

Name of Institution

We have taken Maruti Suzuki as a company to analyze and interpret sales data and impact of time or segment on total sales of the company. Sales figures are of 3 months July, September, August.

Findings Were:
F [ Cal ] is more than F [ Tab ] @ 0.05 Alpha. Outcome was 5.94 where as table value was 4.45 As per Fisher Test Rule if F [ Cal ] is Greater that means Alternate Hypothesis is accepted i.e Ha = 1 2 3 .. k.

Calculations

Name of Institution

Calculation Cont..
SSE = SST- SSTR SSBL SSE = -153513450 Also, MSTR = SSTR/k-1 = 113993745 MSE = SSE/(k-1)(b-1) = -19189931 F(cal)= MSTR/MSE = 5.94 For F(table) value, Numerator D.F = k-1 = 2 Denominator D.F = (k-1)( b-1) = 2*4 = 8 Where, k = number of treatments b= number of blocks

Name of Institution

Conclusion

Name of Institution

At 0.05 level of significance, the above calculated matrices which Shows F Which has come out to be 5.94 where as table value shows F = 4.45 Hence it is proven that Alternate Hypothesis is accepted. Which defines there is no significance and interrelation between sales of different segment cars in different months. Ha = 1 2 . k As we have found out there is no significance of time or month in sales of a vehicle or any segment, Having said that as a manager of company we can introduce any car of any segment at any point of time in a year. However in real market scenario sales fluctuations do happen because of various reasons that have to be monitored as a Company.

Conclusion Cont.

Name of Institution

Factors such as: Festive Season - People tend to purchase more considering the fact auspicious occasions will bring prosperity with the vehicle Eg. Holi, Diwali , Eid such festivals. Economic conditions Interest rates hike which leads to a costlier vehicle. High input cost or rather increasing costs pushes manufacturers to increase the prices Eg. Recently Maruti Suzuki increased prices of its cars up to 10000/-

Name of Institution

## Recommendations and Reference

Recommendations:

Companies should keep a close watch on market trends and customer requirements, Thereby keep making modifications to be in demand. Try to capitalize maximum of customer Emotional Quotient by making such ads which creates huge impact on buyer.

Reference:
http://www.marutisuzuki.com/ Book By: Quantitative Methods for Business David R Anderson

Name of Institution

Thank You !