==Cluster Analysis for TFT Display Mobile Phones==
We performed Cluster Analysis on the present mobile phones available in the market. The purpose of the study was [[Image:cluster analysis.jpg]]
:* To identify the appropriate cluster for the launching of the AMOLED mobile phones under smartphone category.
:* To understand the present positioning of the leading smartphone manufacturing companies.
* '''Clustering Method :''' We applied '''K-means''' clustering method to identify different clusters for the sample. K-means clustering has the researcher specify the number of clusters in advance, then the algorithm calculates how to assign cases to the K clusters. K-means cluster analysis uses Euclidean distance. Initial cluster centers are chosen randomly in a first pass of the data, then each additional iteration groups observations based on nearest Euclidean distance to the mean of the cluster. That is, the algorithm seeks to '''minimize within-cluster variance''' and '''maximize variability between clusters''' in an ANOVA-like fashion. Cluster centers change at each pass. The process continues until cluster means do not shift more than a given cut-off value or the iteration limit is reached.
: [http://faculty.chass.ncsu.edu/garson/PA765/cluster.htm Source: Statnotes, North Carolina State University]
* '''Samples used for Clustering:''' Samples consist of leading players in the smartphone mobile category. These players were identified with reference to Gartner reports.The sample consist of smart phones announced in last three years. They are primarily the mobile phones available in the North America. After aggregating the data we used K-clustering method to identify the different clusters for the samples. We divided our samples in two parts - '''Candybar Design''' and '''Clamshell/Slider Design''' - with thickness being the differentiating parameter. This was done in order to eradicate the error, due to the design of the phone. Our study revealed '''5 clusters''' in the candy-bar mobile handset category and '''4 clusters''' in the clampshell-slider category. * '''Parameters considered for defining clusters:''' The key parameters that were considered for cluster analysis were: :* '''Camera''' (megapixels):* '''Input''' (normal keypad, keyboard and touch-screen):* '''Display Size''':* '''WLAN''' :* '''Colour''':* '''Resolution''':* '''Thickness''' :* '''Price''' We have identified three key factors - '''Input, Display Size and Price''' - that would determine if a cluster would move towards OLED or not. The rationale behind this selection is that Input and Display Size are related to the viewing aspect and these high end mobile phones being high priced, the consumer is expected '''not''' to be price-sensitive. Out of those clusters, for comparison with OLED projections based on analogous products, we considered '''Apple iPhone ''' because of constraint of availabilty of sales data for other models like Blackberry, Nokia N96, HTC Dreamers etc.
