Fixing typos and small issues from bug bash

Technical Deployment Guide/TechnicalDeploymentGuide.md

@@ -43,8 +43,9 @@ The solution architecture consists of the following Azure components:
 * **Azure SQL DB**, used to store several different types of data, pre-process the transactional data for modeling,
   and generate pricing suggestions. A premium edition (P1) is recommended as the larger tables take advantage of clustered columnstore indices.
 * **Azure Storage** account, used to save the model and intermediate data in **Blobs**.
-* A model build **AzureML web service**, running in batch mode, 
-* A collection of several interactive **Azure ML services** for querying the model
+* A model build **AzureML web service**, running in batch mode. 
+* A collection of several interactive **Azure ML services** for querying the model. All services are created
+  by turning the pricing engine python package into a web service on the AzureML platform.
 * A **PowerBI dashboard**, hosted in a **Azure Web App**
 * **Azure Data Factory** for scheduling regular execution
 

User Guide/UserGuide.md

@@ -153,10 +153,10 @@ First, we need to build a model from transactional data. You can either use the
 or input your own data in provided [the spreadsheet](https://aka.ms/pricingxls). Please download
 the spreadsheet.
 
-**Pre-built model.** Out-of-the-box, the solution database is pre-populated with the demonstration dataset 
-(Orange Juice) with dates shifted to create a realistic appearance of data coming in weekly. 
+**Pre-built model.** The solution database is pre-populated with a demonstration dataset 
+([Orange Juice](https://www.rdocumentation.org/packages/bayesm/versions/3.0-2/topics/orangeJuice)) 
+with dates shifted to create an appearance of data coming in weekly. 
 The pre-configured model is automatically re-built weekly from the current data.
-
 You will therefore have a model called 'latestDemoBuild' available out of the box.
 
 When the implementor connects the solution to your business data warehouse, data updates 
@@ -248,6 +248,8 @@ Note we have filtered down to only the elasticities estimated in the latest mode
 Each model run estimated elasticities from the entire history up to and including the model run date.
 With long histories, models tend to be stable: if you click through the model runs, the results are very similar. 
 
+#### Optional: Acessing elasticities from Excel
+
 You can also query the elasticities using the Excel service GetElasticities.
 With this service, you can query the latest model build (use "latestModelBuild" for datasetName).
 Because elasticities can change in time, a query date is required in cell A6 of the spreadsheet.
@@ -385,4 +387,4 @@ ounterfactual is visualized as trend line.
 ## References
 
 1. [Blog post on an early version of the Pricing Engine](https://blogs.msdn.microsoft.com/intel/archives/1015)
-2. [Wikipedia: Price Elasticity of Demand](https://en.wikipedia.org/wiki/Price_elasticity_of_demand)]
+2. [Wikipedia: Price Elasticity of Demand](https://en.wikipedia.org/wiki/Price_elasticity_of_demand)

Walkthrough 1 - Promotion Simulation/PromoSimulationWalkthrough.md

@@ -29,9 +29,13 @@ If you closed that page, you can come back to it from the
 In the spreadsheet, find the Simulate Promotion tab. It has a yellow box with instructions.
 You have already completed the first step of those instructions (set up web service).
 
-Follow the instructions, but choose A18 as the destination cell. 
+Follow the instructions in the spreadsheet, but choose A18 as the destination (output) cell. 
 As <tt>datasetName</tt> parameter of the service, you can use 'latestDemoBuild'
 if the solution has run for a while and a model has been generated.
+For forecastPeriod, choose 3. 
+
+If the table at A18 fills out successfully, then your model has already been built by Azure Data Factory. 
+If you get an error, please build a model according to the instructions in the next section.
 
 Experiment with the price of Minute Maid on week 2. Note that both the sales
 of tropicana and dominicks are affected the same week (cannibalization)