Data Analysis & Forecasting: Wind Speed using Linear Forecasting

This report focuses on analyzing wind speed data using various statistical techniques and linear forecasting models. The analysis includes organizing data into tables and charts, calculating median, mean, mode, range, and standard deviation. A linear forecasting model is developed to create a regression equation, and the values of 'c' and 'm' are computed. The report also forecasts wind speed for the 11th and 13th days using the derived equation. The conclusion emphasizes the importance of these analytical functions in understanding wind speed patterns and forecasting future values. Desklib is a platform where students can find similar solved assignments and past papers for their studies.

DATA ANALYSIS
TECHNIQUES

Table of ContentsINTRODUCTION.......................................................................................................................3
TASK........................................................................................................................................3
1. Organise the information in a format of table................................................................................3
2. Develop charts with the help of above tabular presentation.........................................................3
3. Carry out calculation by describing all the steps included during computation..............................4
4. With the help of linear forecasting model develop the regression equation and compute the
value of c and m.................................................................................................................................5CONCLUSION...........................................................................................................................7
REFERENCES.....................................................................................................................................8

INTRODUCTION
The report prepared takes in account the collection of data in specific areas such as
windspeed. The information in report prepared helps in carrying out analysis which would help
to perform various calculations. Therefore, the regression equation will be developed and the
forecasting related to computation of 2 days would be carried out on the basis of equation formed
(Dhanya and Harish, 2018).
TASK
1. Organise the information in a format of table.
Day
Wind
Speed
1 25
2 27
3 30
4 32
5 23
6 13
7 17
8 17
9 13
10 15
Total 212
2. Develop charts with the help of above tabular presentation.

3. Carry out calculation by describing all the steps included during computation.
Median: It can be defined as a figure which is derived with assistance during collection of data in
ascending order. It can further be counted as mid value among the series formed and that term is
denoted as median (Li, Guo and Zhou, 2021).
Steps followed for calculating median:
Step 1: At first place arrange the information collected in form of ascending.
Step 2: After that compute the number of observation if it results to be even or odd.
Step 3: If it results to be even, then the stated formula can be used (n/2).
Step 4: If it results to be odd then the given formula must be used (n+1/2).
Step 5: The recorded result or output is the position of median.
Median = (N+1 / 2)
(10+1) / 2
11/2 = 5.5
Median of wind speed are as follows:
25, 27, 30, 32, 23, 13, 17, 17, 13, 15
13, 13, 15, 17, 17, 23, 25, 27, 30, 32
Hence the median in this information is (10 + 1) / 2
Median = 5.5th position
Median = (17 + 23) / 2
Median = 20
1. Mean: It can be denoted as an average digit or figure for which the information is
being sorted. From above data the average of windspeed is being computed and the
data is for 10 consecutive days.
Steps involved in calculation of mean:
Step 1: Determine the figures given.
Step 2: Sum all the values which are being laid down.
Step 3: Total the number of observations.
Step 4: Divide the summed-up observation with total number of observations.
Mean of Wind Speed = Sum of data set / Total number of data set
Mean = 212 / 10
Mean = 21.2

Mode: It can be explained as value which results on a repetitive basis in data series being
prepared (Obasi and Shafiq, 2019).
Steps involved in computation of Mode:
Step 1: Accumulate and organise the information given.
Step 2: Assess the distinctive values.
Step 3: Find the frequency of data being observed to occur.
Step 4: Most occurring or repetitive value is said to be mode.
13, 13, 15, 17, 17, 23, 25, 27, 30, 32
Mode = 13 and 17
2. Range: It can be explained as a method or function which helps to compute difference
or variation between highest or lowest amount.
Steps involved or followed in computation of range:
Step 1: Sort the information available.
Step 2: Find which among the series formed is the highest and lowest value.
Step 3: Deduct the least digit from the highest.
Step 4: The value incurred after 3rd step is Range.
Range = Highest digit – Lowest digit
= 32 – 13
Range = 19
Standard deviation: It explains the deviation observed among the series being built and prepared
for the data of windspeed (Wong and Khor, 2019).
Steps needed for carrying out calculation of Standard deviation
Step 1: At first place find the mean of information given.
Step 2: For every observation assess the variation between the value and the mode of
information.
Step 3: Add all the values from step 2.
Step 4: Divide by number of terms present (n).
Step 5: At last, square root the result from step 4.
Standard Deviation= √∑ (xi – μ) 2 / N
=√453.6 / 10
Standard Deviation = 45.36
4. With the help of linear forecasting model develop the regression equation and compute the
value of c and m.
Linear forecasting theory: Such theories are developed on the base of previous prepared
computation report (Paramesh and Shreedhara, 2019). It is further useful in carrying out future
forecasting of results
Following are the steps which is requires in Linear forecasting theory:
Step 1: Search the issue.
Step 2: Find and sort the data.
Step 3: Observe the examination at beginning stages.

Step 4: Pick the relevant models keeping future based practices in mind.
Step 5: Check all information and assess performance given by chosen or selected model.
Y = mx + c
Where ‘y’ denotes dependent factor,
‘mx’ refers to independent factor
‘c’ reflects constant factor.
a) Compute the quantity of m by explaining the steps of computation.
Step 1: Multiply the terms.
Step 2: Compute the sum of total observation.
Step 3: Add each value on a separate basis.
Step 4: Apply multiplication on both elements.
Step 5: At last compute the value.
M = (10*1012 – 55*212) / (10*385 – 3025)
= (10120 – 11660) / (3850 – 3025)
= -1540 / 825
m = -1.87
b) Figure the value of c by jotting down the steps of calculation.
Step 1: Compute the sum or value of ‘y’ variable.
Step 2: Search the value of ‘m’.
Step 3: Compute the summation of ‘x’ variable and afterwards multiply both ‘m’ and
the sum of ‘x’ as well.
Step 4: Subtract summation of ‘mx’ from sum of ‘y’.
Step 5: Afterwards divide the value of ‘mx’ from sum of ‘y’.
C = (212 + 1.87 * 55) / 10
c = (212 + 102.85) / 10
c = 314.85 / 10
c = 31.49
c) Compute the value for m and c on 11th and 13th day.
Windspeed on Day 11:
m = -1.87, c = 31.49, x = 11
y = mx + c
= -1.87*11 + (31.49)

= -20.57 + 31.49
y = 10.92
Wind Speed on Day 13:
m = -1.87, c = 31.49, x = 13
y = mx + c
= -1.87 * 13 + (31.49)
= -24.31 + 31.49
Y = 7.18
CONCLUSION
From above developed report it can be asserted that the information of windspeed is
analysed on the basis of median, mode, mean, standard deviation and range as well. The
regression equation can be prepared with the assistance of forecasting model. It also helps to
understand the importance and necessity of such functions.

REFERENCES
Books and Journals
Dhanya, N.M. and Harish, U.C., 2018. Sentiment analysis of twitter data on demonetization
using machine learning techniques. In Computational vision and bio inspired
computing (pp. 227-237). Springer, Cham.
Li, H., Guo, S. and Zhou, H., 2021. In-situ/operando characterization techniques in lithium-ion
batteries and beyond. Journal of Energy Chemistry. 59. pp.191-211.
Obasi, T. and Shafiq, M.O., 2019, December. Towards comparing and using Machine Learning
techniques for detecting and predicting Heart Attack and Diseases. In 2019 IEEE
international conference on big data (big data) (pp. 2393-2402). IEEE.
Paramesh, S.P. and Shreedhara, K.S., 2019. Automated IT service desk systems using machine
learning techniques. In Data Analytics and Learning (pp. 331-346). Springer, Singapore.
Wong, S.F. and Khor, S.M., 2019. State-of-the-art of differential sensing techniques in analytical
sciences. TrAC Trends in Analytical Chemistry. 114. pp.108-125.