Market Share Forecasting Using Markov Processes
Uploaded by
VISHHAL GUPTAMarket Share Forecasting Using Markov Processes
Uploaded by
VISHHAL GUPTACommon questions
Powered by AITo estimate the transition matrix, use the changes in market share percentages over the given periods. The observed market shares are: Period 1: 50% and 40%, Period 2: 60% and 40%, Period 3: 70% and 30%. By observing changes across periods and assuming a Markov process, calculate transition probabilities. For example, if product share increased from 50% to 60% between periods 1 and 2, the transition matrix elements can be derived using changes: \[T_12=\frac{60-50}{(50)}=0.2\]. Repeat for all periods and ensure row sums equal 1. Interpreting significance involves assessing stability and prediction accuracy for market shares using the matrix .
Revising the transition matrix depends on evaluating the deviations between the predicted and actual period 4 market shares. Given the previous periods' steady increase and assuming a matrix derived from them, a small deviation may not necessitate changes. If the deviation of actual from predicted shares (say, predicted 70% vs. actual 71%) is minor and within expected variations, revision may be unnecessary. However, significant discrepancies indicate model inaccuracies or external factors affecting stability, warranting reevaluation of transition probabilities to reflect new trends .
To calculate the long-run (steady-state) distribution of payment schemes, find the eigenvector of the transition matrix that corresponds to an eigenvalue of 1. The transition matrix is: \[ [0.85, 0.10, 0.05], [0.04, 0.90, 0.06], [0.02, 0.23, 0.75] \]. Solve for the vector \(v\) where \(Tv = v\), subject to \(v_1 + v_2 + v_3 = 1\). Solving yields the long-run distributions approximately: \(v_1 \approx 0.607\), \(v_2 \approx 0.218\), \(v_3 \approx 0.175\). Thus, the long-run percentages are 60.7% for scheme (1), 21.8% for scheme (2), and 17.5% for scheme (3).
To forecast BA's market share after two years, you need to model the changes using a Markov process with the given transition probabilities. The initial market share of BA is 40%, and the competition holds 60%. The transition matrix is: Last flight \ Next flight \ (BA, Competition): (0.90, 0.10), (0.20, 0.80). Calculate the market share after each transition by multiplying the initial state vector [0.40, 0.60] by the transition matrix for two periods. After one year (one transition): [0.40, 0.60] * [ [0.90, 0.10], [0.20, 0.80] ] = [0.46, 0.54]. After the second year: [0.46, 0.54] * [ [0.90, 0.10], [0.20, 0.80] ] gives [0.514, 0.486]. Therefore, BA's market share is forecasted to be 51.4% after two years .
To predict the distribution after two quarters, use the initial state vector of payment schemes [0.60, 0.30, 0.10] and the given transition matrix: \[ [0.85, 0.10, 0.05], [0.04, 0.90, 0.06], [0.02, 0.23, 0.75] \]. After one quarter: Multiply the initial vector by the matrix to get [0.603, 0.270, 0.127]. For two quarters, multiply again: [0.603, 0.270, 0.127] * transition matrix, resulting in approximately [0.616, 0.245, 0.139]. Therefore, the predicted distributions are 61.6% for scheme (1), 24.5% for scheme (2), and 13.9% for scheme (3) after two quarters .