Scribd Logo
Upload

Welcome to Scribd!

  • 36 views

    Exponential Trend Model

    Uploaded by

    mehdi
    The document describes exponential and logistic trend models for forecasting demand. The exponential model assumes demand increases proportionally over equal time periods, while the logistic model accounts for demand initially growing moderately, picking up, and ultimately reaching saturation. The document also provides examples of using linear regression and maximum likelihood estimation to calibrate the models to historical data and forecast future demand.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Exponential Trend Model

    Uploaded by

    mehdi
    36 views20 pages
    The document describes exponential and logistic trend models for forecasting demand. The exponential model assumes demand increases proportionally over equal time periods, while the logistic model accounts for demand initially growing moderately, picking up, and ultimately reaching saturation. The document also provides examples of using linear regression and maximum likelihood estimation to calibrate the models to historical data and forecast future demand.

    Original Description:

    5852

    Original Title

    CE415_LEC8

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document describes exponential and logistic trend models for forecasting demand. The exponential model assumes demand increases proportionally over equal time periods, while the logistic model accounts for demand initially growing moderately, picking up, and ultimately reaching saturation. The document also provides examples of using linear regression and maximum likelihood estimation to calibrate the models to historical data and forecast future demand.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    36 views20 pages

    Exponential Trend Model

    Uploaded by

    mehdi
    The document describes exponential and logistic trend models for forecasting demand. The exponential model assumes demand increases proportionally over equal time periods, while the logistic model accounts for demand initially growing moderately, picking up, and ultimately reaching saturation. The document also provides examples of using linear regression and maximum likelihood estimation to calibrate the models to historical data and forecast future demand.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 20

    Exponential Trend Model

    Exponential
    The increase in demand in equal time periods is not
    constant. The increase in demand, (Pn+1 Pn ), is
    proportional to the present demand, Pn
    i.e., Pn+1 Pn = Pn; where, is the growth rate.
    P1 P0 = P0 or P1 = (1+ )P0
    P2-P1 = P1 or P2 = (1+ )P1 = (1+ )2P0

    Pn = (1+ )nP0
    and P0 can be calibrated by least square regression
    method after linearisng the above equation.
    Logistic Trend Model
    The air travel demand initially grows moderately, picks up when the
    economic base reaches a certain minimum level, and ultimately reaches a
    saturation level. This cycle of demand growth pattern is best depicted by a
    logistic model.
    The rate of growth (t), as in exponential model, is not constant, but is a
    linearly decreasing function of the demand level (Pt).
    i.e., t = a b Pt (1)
    1 dPt
    but, t = (2)
    Pt dt

    therefore, dPt (3)


    = Pt (a bPt )
    dt
    The solution to the above differential equation is the logistic model
    1
    Pt =
    1 b at b (4)
    e +
    P0 a a

    From (1), at t = , t = 0 and therefore, P = a/b


    Calibration of Logistic Trend Model
    Generalized least squares or maximum likelihood method
    can be used to calibrate the logistic model.
    Alternatively, the parameters a and b can be estimated by
    plotting between (Pn+1 Pn)/Pn and Pn. The intercept is a
    and the absolute value of the slope is b.
    (Pn+1-Pn)/Pn

    b
    a 1

    Pn
    Typical Trend Models

    200.00
    Observed Linear Exponential
    Passengers in Lakhs

    150.00

    100.00

    50.00

    0.00
    1970 1975 1980 1985 1990 1995 2000 2005 2010
    Year
    Comparative Models
    These models are also known as
    Ratio methods
    Market share models
    Typical Comparative Models
    A N
    PT = K A PT
    A N
    PT = K A ( PT b)
    A
    PT = K P B T t
    Trend of Proportion of traffic

    Mumbai Delhi Chennai Kolkata


    Proportion of Pass. Traffic

    0.6
    0.5

    0.4
    0.3
    0.2

    0.1
    0
    1970 1975 1980 1985 1990 1995 2000
    Year
    A Top-Down Approach
    Aggregate Forecast for the Nation by
    Econometric/Trend Model
    Allocation to specific region based on historical
    share/economic growth
    Airport choice model for allocation to airports
    with in the metropolis
    Estimation of Induced traffic because of hub
    facility by market surveys/stated preference
    surveys
    Refine forecasts based on experts opinion
    Econometric Trend Models
    Relate Economic and Social Factors to
    Aviation Activity
    Trip generation Models
    -Multiple Linear Regression Technique
    Distribution Models
    - Gravity Model
    Choice Models
    Explanatory Variables used in
    Econometric Models
    Size and Traffic Potential
    - Population
    - GDP
    - Industrial production
    - Per capita income
    - Personal expenditure
    - Leisure time
    - Interregional Linkages:
    Economic and Social
    Transportation
    - Accessibility
    Distance to airport, Travel time to airport
    - Competition
    relative cost, relative travel time,
    schedule and reliability of alternate modes
    - Cost of air travel
    average fare, total travel cost and
    value of travel time
    - Schedule convenience
    service frequency, time of departure,
    necessary connections
    - Service reliability
    on time performance, Cancellation history
    - Transport time
    airport-to-airport time, door-to-door time
    Typical Econometric Models

    Domestic air travel at Frankfurt Airport


    Yt = -1.5298 + 0.61GDPt
    (1.877) (6.637) R2 = 0.82

    International air travel at Frankfurt Airport


    Yt = -15.754 + 2.607GDPt
    (-6.316) (9.765) R2 = 0.87
    Statistical Testing
    Coefficient of determination
    (Yest Yav )
    2

    R2 =
    (Y Yav )
    2

    Standard Error of Y estimate


    0. 5
    (Y Yest ) 2
    SE =
    m (n + 1)

    t-statistic
    t-value = Parameter estimate/Standard error of parameter
    Example Problem
    Year Enplaned Populatio Per capita
    Passengers n of the disposable
    (millions) city, income of the city,
    millions $
    City Region
    1976 0.5 94 0.125 8500
    1979 0.8 135 0.160 9500
    1982 1.00 165 0.200 10500
    1985 1.20 185 0.222 11000
    1988 1.30 200 0.235 12000
    1991 1.45 220 0.255 13500

    Develop a top-down forecast for the 1997 enplaned passengers at the city airport by
    fitting a suitable trend model for the region.
    Develop an econometric model of enplaned passenger growth at the city airport. Also
    prepare a bottom-up forecast of the enplaned passengers at the city airport in 1997 if it is
    expected that the percapita disposable income will be $15,000 and the population will be
    310,000.
    Passenger Traffic in Important Airports

    Mumbai Delhi Chennai Kolkata


    140
    120
    Passengers in Lakhs

    100
    80
    60
    40
    20
    0
    1970 1975 1980 1985 1990 1995 2000
    Year
    Forecasting Model for Mumbai
    For Mumbai airport, an exponential relationship between
    economic growth and traffic was assumed:

    Tt = Tt-1 [1 + (Elasticityt  GDPt)]

    Where,
    Tt = Forecasted passenger traffic level in
    Mumbai in year t
    GDPt = Forecasted GDP growth rate (in decimal) for
    India in year t
    Elasticityt = Elasticity of international or domestic traffic
    growth with respect to GDP growth at
    Mumbai in year t
    Parameters Used in Mumbai Model

    For passenger traffic, the following elasticity values were used:


    2004-2015: Gradually increasing from 1 to 1.3
    2016-2030: Annually decreasing with 0.04 per year
    2031 and beyond: 0.7
    For cargo, a similar model is used, only with different values for the
    elasticity:
    2004-2015: Gradually increasing from 1 to 1.9
    2016-2030: Annually decreasing with 0.04 per year
    2031 and beyond: 1.3
    Passenger Forecasts for Mumbai Airport
    Cargo Forecasts for Mumbai Airport

    You might also like