Demonstration Properties for the TAUREAN Residential Valuation System

Demonstration Properties for the TAUREAN Residential Valuation System Taurean has provided a set of four sample subject properties to demonstrate many of the valuation system s features and capabilities. The properties came from real estate ads published in the Milwaukee Journal Sentinel on Sunday, Aug. 5, 2001. The values and numbers reported in this write-up are current as of Aug. 5, 2001. Numbers may change as they can be affected by future sales updates which occur on a regular basis to maintain the Taurean database with a current reflection of market activity and performance. Access Taurean s web site at: www.taurean.com Within the Registered Users section, enter your assigned User ID and Password in the appropriate fields. Click Login. Make sure that the default Action is to Perform an e-valuation. Click Go. Analysis of Subject 1 Address 4167 S Lenox Ave, Milwaukee WI Sale Price - $105,000 Sale Date July 17, 2001 ID the Subject Enter the following input parameters: State = Wisconsin County = Milwaukee Street Number = 4167 Street Name = le (only enter the first few characters) Search Results Click the address for the correct subject property = 4167 S LENOX ST. Subject Attributes Review the Subject Attributes currently available in Milwaukee County. In particular, note those attributes that can and cannot be modified, as well as those attributes that can be filtered according to minimum and maximum values. Click Attic Condition to review the choices. 1

Analysis Parameters Date Range From Date the furthest date back in time from the To Date, which is the As-Of-Date for the analysis. Default is 12 months back from the As-Of-Date. To Date the As-Of-Date for the analysis. Geography Change to 07/10/2001. This will change the As-Of-Date to 6 days prior to the sale date of the subject. Census Tract default option. Census Block Group small neighborhood unit. Radial Search Review options in the drop-down list. Test Value Estimate Will be demonstrated later. File No. General reference field for the user. Report Type Controlled by Access Level. Number of Most Similar Sales Used to set the table of Most Similar Sales. The default is 10, and the maximum is 15. Do Not Make Any Changes QuickCount Displays the number of comparable sales meeting the Date and Geographic search criteria specified above. Click Quick Count and the system should show 20 properties found with last query. Click Continue to perform the valuation. This will take a few seconds. TRVS Report The following Options are listed horizontally: Printer-Friendly Report Comparable Selection Appraisal Forms & Addenda 2

Note the following sections of the TRVS Report: Input Summary (upper left corner) Pay particular attention to Subject Attributes and Attribute Filters. This is a summary of what the user has changed in the subject property inventory and any attribute filters that may have been activated. For the subject property being processed, the Subject Attributes should indicate No Changes and the Attribute Filters should indicate No Filters Used. Valuation Summary (upper right corner) The Value Estimate is the final estimate derived for the subject. The Value Range is the minimum and maximum adjusted sale prices from the final subset of comparables used to price the subject; the detail of the adjusted sale prices is located in the Most Similar Sales section. The Composite Quality Score is the overall judgment of the supportability of the Value Estimate; the detail is located in the Quality Score Summary section. The Number of Comparable Sales is the number of sales used to evaluate the subject property; the detail is located in the Available Sales Price Distribution section. Finally, the comparable addresses used to price the subject property are listed in order of similarity to the subject. For the subject property, the Value Estimate should be $104,500, with a Value Range of $89,382 to $113,441, a Composite Quality Score of 7, and Number of Comparable Sales equal to 20. Other Subject Property Information This section will display any extra informational data concerning the subject or its location, such as Parcel Number, Census Block Group, and Neighborhood Code. Unusual Characteristics This section identifies any subject property attributes that appear to be unusual when compared to corresponding attributes of the comparables. This is a statistical-based evaluation that compares each subject attribute value to expected values, determined from the distribution of attribute values from the comparables. In default mode, the expected value is targeted to be within the value range approximately 85% of the time (a standard deviation score of 1.75). If any attribute of the subject falls either below or above the expected range, the attribute is listed in this section. This allows the user to easily determine if the subject property exhibits attributes that are unusual and thus may require additional review. Any attribute that is listed in this section should be reviewed closely in the Final Comparable Sales Attribute Listing and Final Comparable Sales Adjustment Detail sections. For the subject being processed, there should be no attributes displayed in this section. Most Similar Sales This section identifies the final set of comparables that will be used to price the subject property. For each identified comparable, the following information is displayed: Selected Comps The rank order of the comparable, if used as a final comparable in pricing the subject. Address Street Number and Street Name. Distance (miles) Distance from comparable to subject property. Sale Price Indicated Selling Price of comparable. Sale Date Indicated Selling Date of comparable. Time Adjusted Sale Price The default analysis permits automatic calculation of an adjustment for time from the comparables that will be used to price the subject property. This allows the micro market of the subject, as represented by the comparables, to be estimated dynamically. 3

Net Attribute Adjustment The net dollar amount of adjustment for difference between the subject and each comparable. Adjusted Sale Price The comparables Time Adjusted Sale Price +/- the Net Attribute Adjustment. The comparables are listed in order of standardized attribute similarity to the subject, with the most similar comparable listed first and the least similar listed last. This ranking is based only on standardized attribute scores prior to attribute differences being converted into dollar adjustment amounts. The final identification of the comparables to use for pricing the subject is based on the dollar adjustment amounts. At the bottom of this section is a Statistical Summary table, which can be very useful to an advanced user. Appraisal theory suggests that as a set of comparables are adjusted first for time and then for differences between themselves and a subject, their aggregate difference should become smaller. If perfect knowledge were available, then the Adjusted Sale Price of each comparable would be the same! However, perfect knowledge is not possible and the best that can be expected is a trend toward smaller variance between Adjusted Sale Prices. For the current analysis, the coefficient of variation under Sale Price is 7.25%, while the coefficient of variation under Time Adjusted Sale Price is 5.80%. The reduction from 7.25% to 5.80% is an indication of how the adjustment for time is moving the sale prices toward a single point estimate. However, the coefficient of variation under Adjusted Sale Price is 7.66%, which is larger than either of the previous estimates. This indicates that the adjustment process, instead of moving the value estimates toward a single point estimate, had just the opposite effect - it made the variance greater! To understand what this means, it is necessary to examine all the entries under the Adjusted Sale Price heading. Reviewing the Adjusted Sale Prices, it can be seen than there is a tendency for the prices to center around $100,000, the exceptions are the last two comparables listed, with Adjusted Sale Prices in excess of $113,000 and another with an Adjusted Sale Price at $89,382. Removing these three transactions, drops the coefficient of variation to 4.58%, which is what is expected. So why are the three questionable comparables available in the first place? The two comparables with Adjusted Sale Prices in excess of $113,00 were the two least like the subject in standardized attribute scores (positions 9 and 10). The default logic of the system is to identify the 10 Most Similar Sales, without regard for their similarity. The last two comparables, while within the top 10 in overall standardized similarity score, were quite different from the preceding 8 comparables. This resulted in selecting a subset of comparable sales in which two of the comparables were possible outliers. The comparable with the Adjusted Sale Price of $89,382 represents a more difficult problem. This questionable comparable is listed third in the list, indicating that it is quite similar to the subject property in standardized similarity score. However, when the units of difference are converted into dollar adjustment amounts, the adjustment amounts tended to be one-sided (negative in this instance), which signify an overall superior condition of this comparable, relative to the subject property. While this condition is not necessarily bad, if this comparable is selected for use in pricing the subject property, it should be reviewed very closely. Quality Score Summary When the analysis is completed, the process is internally reviewed and scored based on its supportability. The characteristics reviewed are based on standard appraisal guidelines and the scoring is based on a 1 (very poor) to 10 (very good) ordinal ranking. The characteristics scored are calibrated such that a score of 7 or greater is generally acceptable. A score of 4 6 is marginal and should be reviewed, and a score of less than 4 should be examined closely before blindly accepting the price estimate. 4

For the subject property being evaluated, the resulting quality scores should be: Price Range = 7 This indicates that the Adjusted Sale Prices of the 10 Most Similar Sales do not vary greatly. Available Sales = 1 This indicates that the initial number of sales was quite small (20) and a larger sample size would have been better. Relationship of Value Estimate = 1 This indicates that the resulting value estimate was not typical for the market (i.e., the comparables used to price the subject) and that the resulting value estimate was either quite small or large, when compared to the market. Higher scores are achieved when the subject property value estimate falls in the middle of the price range of the comparables. Distance to Comparables = 10 This indicates that comparables used to price the subject are very close to the subject property. Inferior/Superior to Subject = 8 This indicates that the balance between some comparables being inferior to the subject and other comparables being superior to the subject is quite good. A low score (1-3) indicates that the subject is either generally superior or inferior to the comparables listed in the Most Similar Sales section, a situation that might suggest further analysis and/or filtering. Quality of Similar Comparables = 9 This indicates that the available generally similar comparables (a subset of the initial 20 and prior to the identification of the Most Similar Sales) is quite good and strengthens acceptance of the limited number of initial comparables. Low scores indicate that even if the initial number of sales is quite large, the number of sales generally similar to the subject is quite small and further analysis and/or filtering may be necessary. Quality of Sales = 10 This indicates that following an attribute-by-attribute comparison between each subject attribute and each corresponding comparable attribute, the overall dollar differences are quite small. A score of 10 in this characteristic is very unusual. A more typical range is 6 to 8 with 7 being typical. This is the single most important Quality Score! Composite Quality Score = 7 Each of the preceding quality scores are weighted and combined into a single overall score. A score of 7 or greater generally is indicative of a very supportable price estimate. Available Sales Price Distribution This is a 10-element histogram of the Time Adjusted Sale Prices of the initial comparables selected. The low and high value range and number of sales within each element is displayed. The price estimate for the subject property is displayed next to the element in which it is located. This can be used as a visual evaluation of the Quality Score Relationship of Value Estimate. For the subject being priced, the $104,500 value estimate is located in the third (3) element, of which there are only two (2) comparables between $103,100 and $107,150. It is always informative to note the elements from which the Final Comparables used in pricing were selected. In this instance, a comparable was selected from the element below the subject price, within the subject price, and above the subject price. Final Comparable Sales Attribute Listing This section lists the attributes of the subject property and Final Comparables used in pricing. In addition this section lists the original subject attribute data, which provides a ready reference to any subject attribute alterations, and any minimum or maximum attribute filters that may have been utilized. At the end of this section is an Adjustment Summary that identifies the Net Adjustment for each comparable, the comparables Adjusted Sale Price and the Subject Property Value Estimate. For the subject being priced, the Net Adjustments were 4,293, -3,469, and 6,996. 5

Final Comparable Sales Adjustment Detail In this section, the attribute listings are divided into the same categories as in the previous Final Comparable Sales Attribute Listing. The purpose of this section is to display the actual dollar adjustment amounts for each attribute for each comparable. This provides the direct market evidence necessary to review and judge the resulting value estimate. Valuation Summary This section summarizes many of the elements presented in previous sections, but it also identifies the Time Adjustment applied. (The default is for the time adjustment to be computed internally from the comparables selected for pricing.) This section also displays Sale Price Statistics identifying the distribution and minimum and maximum values of the sale prices. For the subject being priced, the time adjustment was 7.00% per year and the Coefficient of Variation of the 20 comparables selected was 11%, indicating a relatively tight distribution of sale prices. TRVS Report Options Printer Friendly Report This option provides a nice print-out of the output report. While this option does not format the report to print on specific page sizes, it does remove the graphics and banners that tend to clutter the final report. To activate: Click Printer Friendly report. Click on your browser s Print icon or click File and then Print from your browser tool bar. Comparable Selection This option provides the user with the opportunity to override the system s selection of Final Comparables used for pricing. This option is based on the premise that the most similar comparables may not be the most acceptable comparables. Situations constantly arise where the application of specific rules (regardless how defined) need to be refined to fit a specific context. For example, if it is necessary to provide the client with a value estimate at or slightly above the client s target value, the comparables automatically selected may need to be adjusted or alternatives used. To assist in this reselection and adjustment, the TAUREAN system provides two options that can help guide the selection process so that professional responsibilities are not violated. To activate: Click Comparable Selection. In addition to previously described comparable data elements, such as Adjusted Sale Price, Address, Distance, Sale Price and Sale Date, this section presents three (3) additional data elements: Selected Order, Comparable Weight, and Similarity Index. The Selected Order identifies the rank order of any selected comparable. This rank order selection (default is 3) is based on the general appraisal rule that the comparables most like the subject should be emphasized and the comparables least like the subject should be de-emphasized. The comparable identified in the first rank order position (1) is the comparable determined to be most like the subject property. The rank ordering of comparables is based on similarity (in dollars) to the subject property. The comparable with the least amount of adjustment is the comparable most like the subject property. 6

The Comparable Weight (%) identifies the measure of importance placed on each selected comparable. The process used to automatically assign measures of importance, as a function of the number of comparables being used to price the subject property, is the sum-of-digits formula. With the default option active, the number of comparables used in pricing is 3 and the sum-of-digits weights for the rank order comparables are 1 = 50%, 2 = 33.3%, and 3 =.16.7%. To view the weights for the first time, click Calculate. Additional user options available are Equal Weight (automatically determined) and Custom Weight (user determined). The Similarity Index is a measure of generalized similarity (in dollars) between the subject and the comparable. The base value for the index is determined during the external calibration process performed quarterly. During the calibration process, the dollar adjustment between each subject being processed and its most similar comparable is captured for later analysis. These values represent measures reflective of current market conditions and the ability of the property inventories to capture the dynamics of the market. A relatively small index base value indicates an ability to capture the dynamics of the market while a large index base value indicates a relatively poor ability to capture the dynamics of the market. However, whatever the base value is, it is representative of the market and can be used to score preceding analyses. During an analysis, each comparable listed in the Comparable Selection table is scored with the base value determined for the county in which the analysis is performed. The range is scaled from 1 to 10 where 10 indicates that a relatively small amount of adjustment was necessary, while a score of 1 indicates a very large amount of adjustment was necessary. The ranges are not equal in scale; it is easier to get lower scores than higher scores. The Similarity Index is provided as a guide in reselecting and/or weighting comparables to be used in pricing the subject property. For example, reviewing the Comparable Selection table, for the subject being processed, it can be noted that Comparable 3 has a Similarity Index of 7 (1 = 10 and 2 = 7). At the bottom of the table, it can be noted that Comparable 10 also has a similarity index of 7, however because it was not selected we know that it is a lower quality 7 than Comparable 3. Due to incomplete property inventories, lack of perfect knowledge regarding what individual market participants are judging, and the general inefficiency of real estate markets, it is not at all unreasonable to identify transactions that are quite similar to the subject in attributes yet are very different in sale price. The Taurean system identifies the most similar comparables, ranks them in order, weights their adjusted sale prices, and estimates the subject property s value. If the user elects to alter these rules, then the user has the responsibility to justify the alteration. Return to the TRVS Report by clicking on TRVS Report. Appraisal Forms & Addenda This option permits the transfer of the Taurean analysis onto a variety of standard appraisal forms. For those who might be interested, try it. 7

Analysis of Subject 2 Address 547 Orchard St, Burlington, WI (Racine County) Sale Price - $127,000 Sale Date July 20, 2001 From the breadcrumb trail located at the top of the screen, click ID the Subject. This will return to the page where you can identify another subject property. ID the Subject Enter the following input parameters: State = Wisconsin County = Racine Street Number = 547 Street Name = or (only enter the first few characters) Search Results Click the address for the correct subject property = 547 Orchard St, Burlington, WI. Subject Attributes Analysis Parameters Date Range To Date the As-Of-Date for the analysis. Geography Change to 07/10/2001. This will change As-Of-Date to 10 days prior to the sale date of the subject. Test Value Estimate File No. Report Type Do Not Make Any Changes 8

Number of Most Similar Sales QuickCount Do Not Make Any Changes Click Quick Count and the system should show 46 properties found with last query. Click Continue to perform the valuation. This will take a few seconds. TRVS Report On the TRVS Report, review the following: Value Estimate = $128,000 (Subject Sold for $127,000) Composite Quality Score = 8 Coefficient of Variation =.0996,.0991,.0892 Moving in the correct direction. Quality of Sales = 10 Final Comparable Sales Attribute Listing Sale condition of subject = 2 Sale condition of comparables = 2, 2, 3 9

Analysis of Subject 3 Address 418 7 th Ave., West Bend, WI (Washington County) Sale Price $134,000 Sale Date July 20, 2001 From the breadcrumb trail located at the top of the screen, click ID the Subject. This will return to the page where you can identify another subject property. ID the Subject Enter the following input parameters: State = Wisconsin County = Washington Street Number = 418 Street Name = 7 (only enter the first few characters, even if the street name is a number) Search Results Click the address for the correct subject property = 418 7 th Av S West Bend, WI Subject Attributes Analysis Parameters Date Range To Date the As-Of-Date for the analysis. Geography Change to 07/10/2001. This will change the As-Of-Date to 10 days prior to the sale date of the subject. Test Value Estimate File No. Report Type 10

Number of Most Similar Sales QuickCount Click Quick Count and the system should show 28 properties found with last query. Click Continue to perform the valuation. This will take a few seconds. TRVS Report On the TRVS Report, review the following: Value estimate = $139,000 (Subject Sold for $134,000) Composite Quality Score = 6 Coefficient of Variation =.1446,.1457,.0737 Moving in the correct direction. Quality of Sales = 4 This is poor quality. Final Comparable Sales Attribute Listing Sale condition of subject = 3 Sale condition of comparables = 2, 4, 5 Summary Need to improve the Quality of Sales score by finding more similar comparables. There are several options that can be examined in attempting to improve the quality of the comparables. Options include changing the From Date (going back further in time), Geography (radial distance), or filtering one or more subject attributes. The review of the initial default analysis suggests that filtering of sale condition may be worth investigating. Note that the subject sale condition is scored 3 (indicating a typical market condition) while one of the selected comparables is scored 5 (indicating that its sale price was considerably above what was expected). For this reason, a value of 4 will be used as a maximum filter value for sale condition. From the breadcrumb trail located at the top of the screen, click Subject Attributes. This will return to the page where you can review the attributes and set filters. Subject Attributes Set a maximum filter value on Sale Condition by entering a 4 in the Maximum Filter Value cell. Analysis Parameters QuickCount Click Quick Count and the system should show 17 properties found with last query. Click Continue to perform the valuation. This will take a few seconds. 11

TRVS Report On the TRVS Report, review the following: Value Estimate = $133,500 (Subject Sold for $134,000) Composite Quality Score = 6 Coefficient of Variation =.1362,.1362,.0583 Moving in the correct direction. Quality of Sales = 5 (Better than previous analysis) This is marginal quality. Final Comparable Sales Attribute Listing Sale condition of subject = 3 Sale condition of comparables = 2, 4, 2 Summary Given the overall quality of available data, (review the Comparable Selection) this is an acceptable analysis. 12

Analysis of Subject 4 Address 1764 N 83 rd St., Wauwatosa, WI (Milwaukee County) Sale Price $160,000 Sale Date July 12, 2001 From the breadcrumb trail located at the top of the screen, click ID the Subject. This will return to the page where you can identify another subject property. ID the Subject State = Wisconsin County = Milwaukee Street Number = 1764 Street Name = 83 (only enter the first few characters, even if the street name is a number) Search Results Click the address for the correct subject property = 1764 N 83 rd St. Wauwatosa, WI Subject Attributes Analysis Parameters Date Range To Date the As-Of-Date for the analysis. Geography Change to 07/10/2001. This will change the As-Of-Date to 2 days prior to the sale date of the subject. Test Value Estimate File No. Report Type Number of Most Similar Sales 13

QuickCount Click Quick Count and the system should show 26 properties found with last query. Click Continue to perform the valuation. This will take a few seconds. TRVS Report On the TRVS Report, review the following: Value Estimate = $156,000 (Subject Sold for $160,000) Composite Quality Score = 4 Coefficient of Variation =.0668,.0600,.0828 Moving in the wrong direction. Quality of Sales = 1 This is unacceptable quality. Final comparable Sales Attribute Listing Sale condition of subject = 3 Sale condition of comparables = 4, 5, 5 Summary Need to improve the Quality of Sales score by finding more similar comparables. Reviewing the Coefficient of Variation for the adjusted sale prices in combination with the Quality of Sales score of 1 suggests that the subject is quite different from the comparables available. To adjust for this condition, the search geography will be changed to allow for a greater number of sales from different areas. The change will be to perform a radial distance search of.75 miles around the subject property. From the breadcrumb trail located at the top of the screen, click Analysis Parameters. This will return to the page where you can change the search geography. Analysis Parameters Geography QuickCount Click the radio button next to Radial Search. From the drop-down menu, click ¾. Click Quick Count and the system should show 55 properties found with last query. Click Continue to perform the valuation. This will take a few seconds. TRVS Report On the TRVS Report, review the following: Value Estimate = $165,500 (Subject Sold for $160,000) Composite Quality Score = 7 Much better than the previous analysis. Coefficient of Variation =.0826,.0807,.0639 Moving in the correct direction. 14

Quality of Sales = 6 This is acceptable quality. Final Comparable Sales Attribute Listing Sale condition of subject = 3 Sale condition of comparables = 5, 3, 3 Summary The comparable pool (Most Similar Sales) seems good, however we would like to shift the price estimate toward target value estimate. Comparable Selection Click Comparable Selection option. Click Calculate to display the comparable weights. Comparable #3 has a similarity index of 3 (the same as #1 and #2). Note that the last comparable in the list also has a similarity index of 3. We will now de-select the current Comparable #3 and select a new #3. Click #3 and delete. Click the empty selected order cell across from the last comparable in the list. Enter the number 3. Click Calculate. Note value change from $165,500 to $162,000 in the upper right corner. Click View. TRVS Report On the TRVS Report, review the following: Value Estimate = $162,000 (Subject Sold for $160,000) Composite Quality Score = 7 Much better than the previous analysis. Coefficient of Variation =.0826,.0807,.0639 Moving in the correct direction. Quality of Sales = 6 This is acceptable quality. Final Comparable Sales Attribute Listing Sale condition of subject = 3 Sale condition of comparables = 5, 3, 5 Summary Only the user can judge the acceptability of the analysis. The importance of moving from the $165,500 estimate to a value closer to the target value of $160,000 is a business decision. 15