Pricing of a house is dependent on various variables (external and internal), for e.g. No of Bedrooms, Area, Location etc.
In our project, we studied some major factors that could be quantified and could affect the pricing of a house: • Price • No. of Bedrooms • No. of Bathrooms • Lot Area • Sq. ft. • Age And we tried to find a relation of these factors with Price of houses.
The U.S. housing market is one that's characterized by what's called "boom-bust" cycles. In a housing boom-bust cycle, prices for housing rise for a time, sometimes steeply, and then decline, also sometimes steeply. U.S. housing prices typically rise and fall, only to rise and then fall again, over and over.
The U.S. housing market went bust beginning in 2006 due to an economic slowdown. As more people became unemployed, their mortgages became too much to afford, and they defaulted. When mortgage defaults increased, lenders began suffering huge losses on mortgage loans they'd given. Lenders then reacted to mortgage defaults by tightening credit, thus making eligible homebuyers scarcer and driving down home prices as a result. Even with millions of homes available for sale, there were relatively few eligible buyers, depressing listing prices even more.
U.S. housing prices are starting to see at least modest annual increases. According to A study in University of California, housing prices may post modest increases of 1 to 2 percent annually through 2020. However, individual markets are expected to experience steeper rises than the expected annual increase of 1 to 2% in housing prices. In housing, location is everything, and prime real estate locations will surly see greater price increases.
Last but not least, real estate markets are cyclical, with boom-bust cycles a regular event, meaning within 15 or so years another real estate boom may occur.
Dynamics of pricing a house in Real Estate market
Study of Real Estate Pricing
DEMAND & REVENUE MANAGEMENT PROJECT
US Housing Market
Collin County Housing Market
Zillow v/s leading Competitors
Q & A
The focus of this project is to understand the dynamics
of pricing a house
We studied major factors that can be quantified and
could affect the pricing of a house
THE U.S. HOUSING MARKET
The U.S. housing market went bust beginning in 2006
The U.S. housing prices may post increases of 1 to 2
percent annually through 2020
COLLIN COUNTY HOUSING MARKET
Resurging Housing market with rapid growth
• Average days on the market
• Month’s supply of inventory
• Increasing trend of average original list prices of homes
• Foreclosures have decreased dramatically over the past three years
Total estimated valuations for 2015 have reached almost $100 billion with 11
percent jump from last year.
• Population has risen
• Collin County’s unemployment rate has reduced (5.9% to 3.3%)
• 42.60% Future job growth (Plano)
• Median household income for Collin County is $81,819
• Median household income for Plano is $95,150
Prices of Substitutes
• Renting becomes more expensive to owning a house
• Zillow rent versus buy calculator: better to buy than rent if
living in home more than 1 year and 11 months
• Super low Mortgage rates
• Expectations of higher prices or rents in the future
• Growth expectations
Choose houses from Zillow
Parameters considered: Property ID, selling price, floor-size-SQFT,
lot-area-SQFT, price/SQFT, number of rooms, number of baths,
Track and collect data in MS Excel for 40 houses over 1 month
Regression Analysis done in BI package: SAS
Test the Hypothesis
Test the accuracy of Regression Equation
• Selling_Price (House) = β1 + β2 *Floor_Size_sqft + β3* Lot_Area_sqft + β4*
Pricepersqft + β5* No_of_Rooms + β6* No_of_Baths + β6* Age_months
• Ho: If there is NO significant effect of independent variables on dependent
β1= β2= β3=0
• H1: If there is any significant effect of independent variables on dependent
β1≠ β2≠ β3≠0
• Ho: The coefficient of the respective parameters is zero.
βi=0 for all i = 1 to 6
• H1: The coefficient of the respective parameters is significantly different from
βi≠0 for all i = 1 to 6
p-value: <.0001 hence, we REJECT our Ho
• Conclusion: There is significant effect of independent variables on
• Proportion of variance in Selling Price that can be explained by the
independent variables: Floor Size, Lot Area, Price per sqft, No. of
Rooms, No. of Baths and Age in months.
Quantitative Effects of the Model
• If floor size increases by 1 sq.ft., Selling Price of the house increases by
• If price per sq.ft. increases by $1, Selling Price of the house increases
• If Age of house was 1 month older then the value of the selling price of
the house increases by $150.
PREDICTIVE ABILITY OF THE MODEL
Estimate as per Regression Equation: $461,549
Selling Price by Zillow: $467,500
Address 7017 Brook Forest Cir Plano, Texas - 75024
Floor Size 3324 ft.
Lot Area 7187 Sq. ft.
Price per sqft $ 141/ ft.
No of Rooms 4 Beds
No of Baths 4 Baths
Prices in Collin County increasing in near future
Exogenous factors affecting Real Estate pricing in Collin County:
• Market Size
• Price of Substitutes
• Customer Expectations
Most Significant factors affecting house price
• Floor Size
• Price / Sq. ft.
• Age of house
97.6 % variance in Selling Price is explained by independent
98.7 % accuracy achieved with our current regression model
ZILLOW V.S LEADING COMPETITORS
9/ 20 times
+/- 5% Range
8/ 20 times
Zillow x x x x x x x
Movoto x x x x x
Trulia x x x x x
Realtor x x x x x x
RealtyTrac x x x x x x x
Include more parameters like garage, Swimming pool etc. to get
more sophisticated Regression Analysis
Combine with macroeconomic and demographic factors to
forecast the price of house.
Comparison between 2 counties of same state or different states
to study which factors impacts more in which area.
Plano is an upcoming market to buy house
• Future Job Market
• High Median Income
• Renting becoming expensive.
What affects the price of house
• Macroeconomic, Employment, Demographic factors
How to estimate the price of house with 98.7% accuracy
Where to find the best estimates of your house