2d:null 2e:["$","title",null,{"children":"Prediction market scoring | Nuño Sempere | Observable"}] 33:Tc53,md`At the end of trading we sum up the trades made by all participants, which reveals the final subsidy paid by the market maker: ${tex`\sum_{t=0}^T (S(a_{t+1})-S(a_t))=S(a_T)-S(a_0)`} (In other words, the market maker only pays the difference between the very first and last trade, as everything in-between is zero-sum among predictors.) Moreover, the first prediction ${tex`a_0`} is submitted by the market maker as a prior, in order to make sure that the first outside trader to enter the market has someone trade against. #### Some intuition Your payment is: ${tex`S(a_{t+1})-S(a_t)`}. This can be interpreted as you buying some shares from the previous trader at a cost of ${tex`S(a_t)`} and selling them to the next trader at a cost of ${tex`S(a_{t+1})`}. (Or, more precisely, all these trades happen through the market maker.) This leaves open the question of how the market maker is to choose an appropriate function ${tex`S(a_t)`}, which I won't cover here. #### Under the hood The above is an instance of a nice isomorphism between share-based prediction markets and forecasting tournaments using scoring rules. The "front-end" (i.e. what participants see) looks like forecasting as usual, but the actual calculations that happen on the "back-end" -- and the incentives that ensue -- are analogous to a prediction market. ### Collaborative markets There are many great things about prediction markets, but they also suffer some flaws. In particular, they are constant-sum. Any improvement accrued by one participant necessarily comes at the cost of someone else (though that someone else might be the market maker!) Sharing useful information can never help me, since it will only allow someone else to execute my good trades before I get the chance. Moreover, if I notice another player being very naive, I can profit by exploiting them -- and I strongly prefer them to *stay* naive so that I can keep profiting from their mistakes. Collaborative prediction markets is a suggestion for alleviating this problem. The intuition is as follows. Consider the dynamics in a well-functioning startup. If someone else makes a mistake, and I have to fix it, that's annoying, and for future I'd rather teach them how to avoid the mistake themselves. That frees us both to work on more important things. Moreover, if I deliberately sabotage things in the company, but in such a way that I look better at the end of the day, that will make things worse for the startup as a whole -- causing all our equity to be worth less. I get a larger proportion of a smaller pie. Similarly, founders often give away equity to investors and employees, thereby taking on a *smaller share* of a hopefully *larger pie*. The idea is that as long as the pie is growing, people are happy to help others even if it reduces their proportional share. So, instead of paying you ${tex`S(a_{t+1})-S(a_t)`}, for each trade, the market maker would pay: ${tex`S(a_T)(S(a_{t+1})-S(a_t))`}. We scale every trade by score of the final aggregate. This means that if you help others perform better trades, this is going to act as a multiplier on your own trades as well. `34:T1270,get_user_scores = async () => { // Collaborative // Data structure to hold usernames and their associated cumulative score let scoresCollaborative = ({}) let numberQuestionPredicted = ({}) let totalPaymentCollaborative = 0 // Iterate over questions let balancesCollaborative = await create_balances_object(true) //question_balances for (const id in balancesCollaborative) { let b = balancesCollaborative[id]["Balance"] // Iterate over users who redicted on that question for (const name in b) { // Check if we've already added an entry for the user, and if not add one if (!(name === "Score_details")) { if (!(name in scoresCollaborative)) { scoresCollaborative[name] = b[name] // gets the score from the user called "name" numberQuestionPredicted[name] = 1 } else { scoresCollaborative[name] += b[name] numberQuestionPredicted[name] += 1 } totalPaymentCollaborative+=b[name] } } } let totalPaymentCollaborativePositive = math.sum(Object.values(scoresCollaborative).filter(n => n>0)); let totalPaymentCollaborativeNegative = math.sum(Object.values(scoresCollaborative).filter(n => n<0)); // NonCollaborative // Data structure to hold usernames and their associated cumulative score let scoresNonCollaborative = ({}) let totalPaymentNonCollaborative = 0 // Iterate over questions let balancesNonCollaborative = await create_balances_object(false) //question_balances for (const id in balancesNonCollaborative) { let b = balancesNonCollaborative[id]["Balance"] // Iterate over users who redicted on that question for (const name in b) { // Check if we've already added an entry for the user, and if not add one if (!(name === "Score_details")) { if (!(name in scoresNonCollaborative)) { scoresNonCollaborative[name] = b[name] // gets the score from the user called "name" } else { scoresNonCollaborative[name] += b[name] } totalPaymentNonCollaborative+=b[name] } } } let totalPaymentNonCollaborativePositive = math.sum(Object.values(scoresNonCollaborative).filter(n => n>0)); let totalPaymentNonCollaborativeNegative = math.sum(Object.values(scoresNonCollaborative).filter(n => n<0)); // Create a dataset suitable for the table. let names = Object.keys(scoresCollaborative) let n = names.length let result = [] for(let i=0; i 0 ? proportionPayoutCollaborativePositive : -proportionPayoutCollaborativeNegative let proportionPayoutNonCollaborative = scoresNonCollaborative[name] > 0 ? proportionPayoutNonCollaborativePositive : -proportionPayoutNonCollaborativeNegative result.push( { name: name, numberQuestionsPredicted: numberQuestionPredicted[name], scoreCollaborative: scoresCollaborative[name], // proportionPayoutCollaborative: proportionPayoutCollaborative, // proportionPayoutNonCollaborative: proportionPayoutNonCollaborative, proportionPayoutCollaborative: proportionPayoutCollaborative, proportionPayoutNonCollaborative: proportionPayoutNonCollaborative, // proportionPayoutCollaborativePositive: proportionPayoutCollaborativePositive, // proportionPayoutCollaborativeNegative: proportionPayoutCollaborativeNegative, // proportionPayoutNonCollaborativePositive: proportionPayoutNonCollaborativePositive, // proportionPayoutNonCollaborativeNegative: proportionPayoutNonCollaborativeNegative, scoreNonCollaborative: scoresNonCollaborative[name], collaborationSubsidy: scoresCollaborative[name] - scoresNonCollaborative[name], collaborationEdge: proportionPayoutCollaborative - proportionPayoutNonCollaborative, } ) } return result }35:T5f0,edges_to_trade_scores = (edges, endCursor) => { // Iterates over edges and isolates aggregates before and after each player prediction (except the first one) // Array to store the output as we proceed let result = new Array() edges.forEach( edge => { let node = edge.node // Check that this is submitted by a player which is not the first player // The COMPETITIVE check is probably unnecessary since I changed the rest of the code to just pull COMPETITIVE measurements from the API, but leaving it in anyway, might be safe to do so if (node.competitorType == "COMPETITIVE" && edge.cursor !== parseInt(endCursor).toString()) { let cursor = edge.cursor // Compute aggregates after and before prediction let a_before = edges.find( e => e.cursor == (parseInt(cursor)+1).toString() ) let a_before_score = a_before.node.measurementScoreSet.nonMarketLogScore let a_after = edge // We just set the aggregate to be whatever trade the last player made. Previous code here: edges.find( e => e.cursor == (parseInt(cursor)-2).toString() ) // Change to -1 to compare with aggregate instead let a_after_score = a_after.node.measurementScoreSet.nonMarketLogScore // Place everything in a data structure to output let output = ({}) output[node.agent.name] = a_after_score - a_before_score result.push(output) } } ) return result }36:Tac3,function createPieChart(isCollaborative, positive, svgA, svgB){ let width= isCollaborative? 300 : 400 let height = 450 let margin = 50 let dataset = getDataSet(isCollaborative, positive) let domElement = isCollaborative? svgA : svgB // "#svg1" : "#svg2" const svg = d3.select(domElement) svg.attr('width', width + margin + margin) .attr('height', height + margin + margin) // let area = isCollaborative ? total_subsidiesCollaborative : total_subsidiesNonCollaborative let area = isCollaborative ? math.sum(dataset.map(object => object.score)) : math.sum(dataset.map(object => object.score)) area = area*300 let radius = Math.sqrt(area / 3.14159265) let colorScale = d3.scaleOrdinal() .domain(dataset) //.range(['#F1892D', '#0EAC51', '#0077C0', '#7E349D', '#DA3C78', '#E74C3C']) .range(['#fffac8', '#800000', '#aaffc3', '#808000', '#000075', '#ffd8b1', '#e6194b', '#3cb44b', '#ffe119', '#4363d8', '#f58231', '#911eb4', '#46f0f0', '#f032e6', '#bcf60c', '#fabebe', '#008080', '#e6beff', '#9a6324']) let totals = dataset.map((element) => element.score); let grandTotal = totals.reduce((a, b) => a + b); let arc = d3.arc() .outerRadius(radius) .innerRadius(0); let pie = d3.pie() .sort((a, b) => d3.ascending(a, b)) .value((d) => d.score) let arcGroup = svg.append("g") .attr("class", "arc-group") .attr("transform", `translate(${margin+radius}, ${height/2 + margin})`); let arcs = arcGroup.selectAll(".arc") .data(pie(dataset)).enter() .append("g") .attr("class", "arc"); arcs.append("path") .attr("d", arc) .style("fill", (d,i) => colorScale(i)); if(!isCollaborative){ let legendPadding = 5; let legendBox = svg.append("g") .attr("class", "legend-group") .attr("transform", `translate(${margin+2*radius+margin}, ${(3/4) * radius})`); let legend = legendBox.selectAll(".legend") .data(dataset).enter() .append("g") .attr("class", "legend") .attr("transform", (d, i) => `translate(${0}, ${i * 20 + 20})`); let displacement = "0.4em" legend.append("circle") .attr("class", "legend-color") .attr("r", displacement) .style("fill", (d, i) => colorScale(i)); legend.append("text") .attr("class", "legend-text") .attr("x", "1em") .attr("y", displacement) .style("font", "14px sans-serif") .style("text-anchor", "start") .text((d) => d.name); } let title = isCollaborative ? "Collaborative" : "Noncollaborative" svg.append("text") .attr("x", margin+radius) .attr("y", 50) .style("text-anchor", "middle") .style("font", "25px sans-serif") .text(title); return svg.node() }37:[] 31:{"id":"7f22791463317c25","slug":"prediction-market-scoring","title":"Prediction market scoring","thumbnail":"040dc5541e33eeda1dc97167a3de074b4a2ef778edf08dc196a53d3db35ef4f8","access_level":"public","create_time":"$D2019-12-04T09:05:14.166Z","update_time":"$D2019-12-06T13:36:38.178Z","view_time":null,"first_public_time":"$D2019-12-04T16:45:39.415Z","has_unpublished_edits":null,"fork":{"id":"3e148a96a190f1ae","slug":"prediction-market-scoring","title":"Prediction market scoring","owner":{"id":"6787ecb194e8d603","login":"jjj","name":"Jacob Lagerros"}},"stars":0,"starred":false,"owner":{"type":"team","id":"07a813a05afbeeaf","login":"nunosempere","name":"Nuño Sempere","avatar_url":"https://avatars-next.observableusercontent.com/512eae7d32fc25bba414910bc3c00fac559004bc68c3e14fa07fa7c6544b630f","tier":"starter_2024"},"version":1468,"publish_version":1468,"role":null,"authored":false,"authors":[{"type":"individual","id":"cee02f8e9acb9aee","login":"nunosempere","name":"Nuño Sempere","avatar_url":"https://avatars-next.observableusercontent.com/bc840e103afe5117def6829a8a8476725da032e7bb7cc3cb8c49257be439a233","tier":null}],"members":[{"type":"individual","id":"cee02f8e9acb9aee","login":"nunosempere","name":"Nuño Sempere","avatar_url":"https://avatars-next.observableusercontent.com/bc840e103afe5117def6829a8a8476725da032e7bb7cc3cb8c49257be439a233","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":1,"collection":null,"type":"notebook","body":{"title":"Prediction market scoring","cells":[{"id":0,"value":"md`# Prediction market scoring\n\nThis notebook implements prediction market scoring for Foretold channels.\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":402,"value":"md`## Overview`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":359,"value":"md`### Results`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":347,"value":"md`Net balance of the house at the end of trading (i.e. total subsidy paid out to predictors):`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":346,"value":"total_subsidiesCollaborative = tableT.map(a=>a.scoreCollaborative).reduce((a,b)=>a+b)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":563,"value":"total_subsidiesNonCollaborative = tableT.map(a=>a.scoreNonCollaborative).reduce((a,b)=>a+b)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":861,"value":"totalCollaborationSubsidy = total_subsidiesCollaborative - total_subsidiesNonCollaborative","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":368,"value":"md` #### Individual payouts:`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":327,"value":"tableT = get_user_scores()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1336,"value":"tableTPositive = tableT.filter(e => e.scoreNonCollaborative >= 0)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1340,"value":"tableTNegative = tableT.filter(e => e.scoreNonCollaborative < 0)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1348,"value":"printTable(tableTPositive)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1355,"value":"printTable(tableTNegative)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":524,"value":"function printTable(T){\n return table(T, {\n sortable: true,\n columns: {\n name: {\n title: \"Name\",\n },\n numberQuestionsPredicted: {\n title: \"Questions Predicted\"\n },\n scoreCollaborative : {\n title: \"Score (C)\",\n formatter: val => val.toFixed(2)\n },\n proportionPayoutCollaborative: {\n title: \"%Σ(Score) (C)\",\n formatter: val => val.toFixed(2)+\"%\"\n\n },\n scoreNonCollaborative : {\n title: \"Score ¬C\",\n formatter: val => val.toFixed(2)\n },\n proportionPayoutNonCollaborative: {\n title: \"%Σ(Score) (¬C)\",\n formatter: val => val.toFixed(2)+\"%\"\n\n }, \n collaborationSubsidy: {\n title: \"Collaboration Subsidy\",\n formatter: val => val.toFixed(2)\n },\n collaborationEdge: {\n title: \"Collaboration Edge\",\n formatter: val => val.toFixed(2)+\"%\"\n }, \n\n\n }\n })\n}\n\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":594,"value":"{ let barCollab = vl.markBar({opacity: 0.5}).data(tableT).encode(\n vl.y().field('scoreCollaborative').type('quantitative').title(\"Score\").scale({type: 'symlog', constant: 0.5}),\n vl.x().field('name').type('nominal').sort(vl.field(\"scoreCollaborative\")),\n );\n let barNonCollab = vl.markBar({opacity: 1}).data(tableT).encode(\n vl.y().field('scoreNonCollaborative').type('quantitative').title(\"Score\").scale({type: 'symlog', constant: 0.5}),\n vl.x().field('name').type('nominal').sort(vl.field(\"scoreCollaborative\")).title(\"Name\"),\n );\n\n \n return vl.layer(barCollab, barNonCollab)\n .width(800)\n .height(500)\n .title(\"Scores: Collaborative (translucent) & NonCollaborative (opaque)\")\n .render();\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":890,"value":"// Uncomment to see the above plot with linear scales, as opposed to symlog scales\n// unalteredScales.render()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":631,"value":"{ let barSubsidy= vl.markBar({opacity: 1}).data(tableT).encode(\n vl.y().field('collaborationSubsidy').type('quantitative').title(\"Score\").scale({type: 'symlog', constant: 0.5}),\n vl.x().field('name').type('nominal').sort(vl.field(\"collaborationSubsidy\")).title(\"Name\"),\n );\n\n \n return barSubsidy\n .width(800)\n .height(500)\n .title(\"Collaboration Subsidy\")\n .render();\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":901,"value":"// Uncomment to see the above plot with linear scales, as opposed to symlog scales \n// unalteredScalesSubsidy.render()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":646,"value":"md `#### Pie Chart\nNote: +1.02 and +0.542 have been added to every collaborative and noncollaborative score respectively, to ensure that each slice has a positive area. This means that JK has area of ~0; if you wanted to visualize him, those numbers ought to be increased.\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1397,"value":"md `### For those with a positive score`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":817,"value":"html ` \n`\n// I am creating two svg elements because I want to render them side by side.\n // They will later be selected and edited by a d3 function (the vega-lite api is not conducive to creating pie charts, as far as I'm concerned.","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1401,"value":"md `### For those with a negative score`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1403,"value":"html ` \n`\n// I am creating two svg elements because I want to render them side by side.\n // They will later be selected and edited by a d3 function (the vega-lite api is not conducive to creating pie charts, as far as I'm concerned.","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":371,"value":"md` ### Payouts by question:`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":188,"value":"question_balances = await create_balances_object(true)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":356,"value":"md`### Settings\n*Tweak these parameters to change the results above*\n`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":386,"value":"md`Set the channel (must be public):`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":377,"value":"channelId = \"f19015f5-55d8-4fd6-8621-df79ac072e15\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":472,"value":"md`Set the collaborative multiplier (since it currently uses log score we want some transformations for it to be positive but not too large)`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":464,"value":"collaboration_multiplier = (final_agg_score) => (final_agg_score+20)/10","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":320,"value":"md`How many questions from the channel to include (lower will make calculation faster...)`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":315,"value":"num_questions = 11 ","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":399,"value":"md`## Explanation\n\n### Market scoring rules\n\nAssume that you join the market at time ${tex`t`}, and that the current aggregate/market price at that time is \n${tex`a_t`}. You make your prediction, which we interpret as a trade that moves the aggregate to ${tex`a_{t+1}`}.\n\nCurrently, the aggregate is set to be the last prediction made. (So the actual average aggregate shown on the platform doesn't matter!) \n\nThe market maker then pays you based on how much you moved the market in the right direction, i.e:`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":429,"value":"tex`S(a_{t+1})-S(a_t)`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":433,"value":"$33","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":396,"value":"md`## Backend`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":292,"value":"md`### Helper functions`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":334,"value":"$34","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":239,"value":"// Creates a dict indexed by question ids, which houses the balances at the end of all questions for all users\ncreate_balances_object = async (collaborative) => {\n \n let balances = ({})\n \n const ms = await measurableIds\n \n for (const measurable of ms) {\n let id = Object.keys(measurable)[0]\n balances[id] = ({})\n let title = measurable[id][\"Title\"]\n balances[id][\"Title\"] = title\n let balance = await get_question_balance(id, collaborative)\n balances[id][\"Balance\"] = balance\n }\n \n return balances\n}","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":116,"value":"// Shows the various trade done by users on a particular question and their profits\nget_question_balance = async (measurableId, collaborative) => {\n\n let final_score = await get_final_score(measurableId) // Get the score of the final trade on the question\n let multiplier = collaborative ? collaboration_multiplier(final_score) : 1\n let balances = ({\"Score_details\": {\"final_score\": final_score, \"final_multiplier\": multiplier}})\n let trade_scores = await trades_from_measurable(measurableId) // Get scores from each trade for each user on the question\n \n // Iterate over trades and add their results to overall balances\n for (const obj of trade_scores) {\n \n let user = Object.keys( obj )[0]\n let trade_score = obj[user]\n \n if (!(user in balances)) {\n balances[user] = multiplier * trade_score \n } else {\n balances[user] += multiplier * trade_score\n }\n \n } \n \n return balances\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":194,"value":"// Gets the score of the final trade for a question\n\nget_final_score = async (id) => {\n let query = `\n {\n measurements(\n measurableId: \"${id}\"\n first: 1\n competitorType:COMPETITIVE\n )\n {\n edges {\n cursor\n node {\n measurementScoreSet {\n nonMarketLogScore\n }\n }\n }\n }\n }\n `\n let response = await request(query)\n let score = response.data.measurements.edges[0].node.measurementScoreSet.nonMarketLogScore\n return score\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":18,"value":"trades_from_measurable = async (id) => {\n let query = `\n {\n measurements(\n measurableId: \"${id}\"\n first: 500\n competitorType:COMPETITIVE\n )\n {\n pageInfo {\n hasNextPage\n startCursor\n endCursor\n }\n edges {\n cursor\n node {\n measurementScoreSet {\n nonMarketLogScore\n }\n competitorType\n agent {\n name\n }\n relevantAt\n }\n }\n }\n }\n `\n let response = await request(query)\n \n let edges = response.data.measurements.edges\n let endCursor = response.data.measurements.pageInfo.endCursor\n \n let scores = edges_to_trade_scores(edges, endCursor)\n \n // TODO: add check to see if question has resolved\n return scores\n\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":41,"value":"$35","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":286,"value":"md`### Data structures`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":100,"value":"measurables_in_channel = async (id) => {\n let query = `\n {\n measurables(\n channelId: \"${id}\"\n first: ${num_questions}\n )\n {\n pageInfo {\n hasNextPage\n startCursor\n endCursor\n }\n edges {\n cursor\n node {\n id\n name\n state\n previousAggregate {\n measurementScoreSet {\n nonMarketLogScore\n }\n }\n }\n }\n }\n }\n `\n let response = await request(query)\n \n let edges = response.data.measurables.edges\n let endCursor = response.data.measurables.pageInfo.endCursor\n \n return edges\n\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":103,"value":"judged_measurables = (await measurables_in_channel(channelId)).filter( m => m.node.state === \"JUDGED\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":108,"value":"// Get resolved questions\nmeasurableIds = judged_measurables.map( edge => {\n let node = edge.node\n let result = ({})\n result[node.id] = ({\"Title\": node.name, \"Final_Score\": node.previousAggregate.measurementScoreSet.nonMarketLogScore })\n return result\n } \n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1146,"value":"function getDataSet(isCollaborative, positiveScores){\n let score = isCollaborative ? \"scoreCollaborative\": \"scoreNonCollaborative\"\n \n let filteredT = positiveScores ? tableTPositive : tableTNegative\n let min = Math.min(filteredT.map(element => element[score]))\n min = min<0 ? -min: 0\n \n let result = []\n filteredT.forEach(element => {\n result.push(\n {\n name: element.name, \n score: positiveScores ? element[score] : - element[score] // previously +min\n }\n )\n })\n return result\n\n \n}","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1420,"value":"// getDataSet(true, true)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1424,"value":"// getDataSet(true, false)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1426,"value":"// getDataSet(false, true)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1428,"value":"// getDataSet(false, false)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":931,"value":"md `### Plot objects`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":944,"value":"\nunalteredScales = { \n let barCollab = vl.markBar({opacity: 0.5}).data(tableT).encode(\n vl.y().field('scoreCollaborative').type('quantitative').title(\"Score\"),\n vl.x().field('name').type('nominal').sort(vl.field(\"scoreCollaborative\")),\n );\n let barNonCollab = vl.markBar({opacity: 1}).data(tableT).encode(\n vl.y().field('scoreNonCollaborative').type('quantitative').title(\"Score\"),\n vl.x().field('name').type('nominal').sort(vl.field(\"scoreCollaborative\")).title(\"Name\"),\n );\n\n \n return vl.layer(barCollab, barNonCollab)\n .width(800)\n .height(500)\n .title(\"Scores: Collaborative (translucent) & NonCollaborative (opaque)\")\n \n}\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":949,"value":"unalteredScalesSubsidy = { let barSubsidy= vl.markBar({opacity: 1}).data(tableT).encode(\n vl.y().field('collaborationSubsidy').type('quantitative').title(\"Score\"),\n vl.x().field('name').type('nominal').sort(vl.field(\"collaborationSubsidy\")).title(\"Name\"),\n );\n\n \n return barSubsidy\n .width(800)\n .height(500)\n .title(\"Collaboration Subsidy\")\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1161,"value":"$36","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":996,"value":"{\n // createPieChart(isCollaborative, positiveScores, svgA, svgB)\n createPieChart(true, true, \"#svg1\", \"#svg2\")\n createPieChart(false, true, \"#svg1\", \"#svg2\")\n\n createPieChart(true, false, \"#svg3\", \"#svg4\")\n createPieChart(false, false, \"#svg3\", \"#svg4\")\n\n return \"Pies created\"\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":288,"value":"md`### Imports`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":14,"value":"import {request} from \"@jjj/bot-tips\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":481,"value":"import {table} from \"@tmcw/tables@513\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":595,"value":"import {vl} from '@vega/vega-lite-api'","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":648,"value":"d3 = require(\"d3@5\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":1386,"value":"math = require('https://cdnjs.cloudflare.com/ajax/libs/mathjs/6.2.2/math.js')","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W37","stdlib":"1","titling":"auto"}} 32:"$31:body"