Objective: Identify factors affect selling price of houses
Details of Data set ::
No of factors : 20
No of Quantitative factors : 10
No of Qualitative factors : 9
No of records : 13580
| Item | Number.Range |
|---|---|
| No of Suburb | 314 |
| No of Council Areas | 34 |
| No of Regions | 8 |
| No of Postcodes | 198 |
| No of Seller Agencies | 268 |
| No of Methods of selling | 5 |
| No of Types of house | 3 |
| Period-House Built | 1830-2018 |
| Period-House Selling | 1/07/2017 - 9/09/2017 |
| Range-Lattitude | (-38.18255) - (-37.40853) |
| Range-Longtitude | (144.4318) - (145.5264) |
| No of records | 13580 |
| No of factors | 20 |
| No of Quantitative factors | 10 |
| No of Qualitative factors | 9 |
| Variable | Min. | X1st.Qu. | Median | Mean | X3rd.Qu. | Max. | NA.s |
|---|---|---|---|---|---|---|---|
| Rooms | 1.00 | 2.00 | 3.0 | 2.938 | 3.00 | 10.00 | NA |
| Price | 85000.00 | 650000.00 | 903000.0 | 1075684.000 | 1330000.00 | 9000000.00 | NA |
| Distance | 0.00 | 6.10 | 9.2 | 10.140 | 13.00 | 48.10 | NA |
| Bedrooms2 | 0.00 | 2.00 | 3.0 | 2.915 | 3.00 | 20.00 | NA |
| Bathrooms | 0.00 | 1.00 | 1.0 | 1.534 | 2.00 | 8.00 | NA |
| car | 0.00 | 1.00 | 2.0 | 1.610 | 2.00 | 10.00 | 62 |
| Landsize | 0.00 | 177.00 | 440.0 | 558.400 | 651.00 | 433014.00 | NA |
| BuildingArea | 0.00 | 93.00 | 126.0 | 152.000 | 174.00 | 44515.00 | 6450 |
| Lattitude | -38.18 | -37.86 | -37.8 | -37.810 | -37.76 | -37.41 | NA |
| Longitude | 144.40 | 144.90 | 145.0 | 145.000 | 145.10 | 145.50 | NA |
| Factors | P.value | Significance | Decision |
|---|---|---|---|
| Intercept | 4.86E-14 | Significant | Affect |
| Land size | 3.61E-05 | Significant | Affect |
| Distance from UBD | 2.16E-05 | Significant | Affect |
| No of Bedrooms | < 2e-16 | Significant | Affect |
| No of Bathrooms | < 2e-16 | Significant | Affect |
| Type of house | < 2e-16 | Significant | Affect |
| Method of selling | 8.43E-11 | Significant | Affect |
| No of carspots | < 2e-16 | Significant | Affect |
| Size of building area | < 2e-16 | Significant | Affect |
| Suburbs | multiple | Significant | Affect |
| Seller | multiple | Significant | Affect |
| Year of built | multiple | Significant | Affect |
| Region Name | multiple | Significant | Affect |
| Councile Area | … | NA | NA |
| Property Counts | … | NA | NA |
Issues:
1. Two columns are number of rooms.
2. Some houses are with 0 rooms.
3. 20 rooms are in a house.
4. More columns contain NAs.
5. More quantitative variables(factors) are inter-correlated.
6. Place related factors/variables are connected/compounding.